From 1d4cffd6ad2f95ad4b6ea059f8668e4fae90b73c Mon Sep 17 00:00:00 2001 From: Nils Date: Tue, 14 Jul 2026 19:44:07 +0200 Subject: [PATCH] integrate LCB dissociation (untrained merge wins far-transfer, p=0.001 vs trained loop), capability panel (no MC damage), L23 exit completes sweep Co-Authored-By: Claude Fable 5 --- PAPER.md | 67 +- results-loop/eval_code_tap23_rerun.json | 3036 +++++++++++++++++++++++ results-loop/eval_lcb_distill.json | 1221 +++++++++ results-loop/eval_lcb_loop.json | 1221 +++++++++ results-loop/eval_lcb_untrained.json | 1221 +++++++++ results-loop/eval_panel_base_k0.json | 8 + results-loop/eval_panel_distill_ff.json | 8 + results-loop/eval_panel_loop_k2.json | 8 + results-loop/fig_transfer.png | Bin 36942 -> 43841 bytes results-loop/lcb_labels.json | 152 ++ scripts/figures_final.py | 8 +- 11 files changed, 6938 insertions(+), 12 deletions(-) create mode 100644 results-loop/eval_code_tap23_rerun.json create mode 100644 results-loop/eval_lcb_distill.json create mode 100644 results-loop/eval_lcb_loop.json create mode 100644 results-loop/eval_lcb_untrained.json create mode 100644 results-loop/eval_panel_base_k0.json create mode 100644 results-loop/eval_panel_distill_ff.json create mode 100644 results-loop/eval_panel_loop_k2.json create mode 100644 results-loop/lcb_labels.json diff --git a/PAPER.md b/PAPER.md index e0aa1b0..e24df21 100644 --- a/PAPER.md +++ b/PAPER.md @@ -34,8 +34,12 @@ degrades them), width rivals depth (trained pause registers reach 36.4%), and verifier-assisted (oracle) sampling wins overall accuracy at matched compute — though the *deployable* selector loses that edge entirely. What survives is precise: the implant owns exactly the plan-dependent slice at -zero visible tokens and zero decode cost, transfers with the substrate -rather than the task, and its placement is dictated by the lens. At 12B a +zero visible tokens and zero decode cost, and its placement is dictated by +the lens. Transfer dissociates by distance: near-distribution the trained +and untrained implants tie (HumanEval); far from it (LiveCodeBench) the +*untrained* merge significantly helps while the trained content +significantly hurts — the learned content is task-local, the recurrence +substrate is general. At 12B a constant merge coefficient destroys the substrate; a state-dependent coefficient (3.8K parameters) restores MBPP but not Blocksworld or GSM8K — the anchor coefficient is the stability dial that unifies this work with @@ -174,8 +178,9 @@ hard 43.6%, overall 53.6%. L13: hard 17.9%, overall 34.4%. L9–L12: overall verified bit-identical) — the 12B model has no shared-KV layers, making it the unconfounded replication. Exit sweep at fixed entrance (L27/30/32/34): hard 39.3–46.4%, within seed spread. The lens boundary is necessary; the -exit is a free parameter. (The L23-exit arm died in training; a rerun is in -progress — the four completed exits bracket it. [L23 PENDING]) +exit is a free parameter — the completed five-point exit sweep +(L23/27/30/32/34) spans hard 39.3–46.4% with L23 at the top (46.4% at k=2), +all within seed spread. ### 3.2 The attribution ladder @@ -297,7 +302,28 @@ significant** (paired McNemar at k=2, 9 vs 7 discordant, p=0.80). What transfers significantly is the *merge perturbation itself*, not the MBPP-trained content — the cleanest evidence that off-distribution value is substrate-shaped rather than task-memorized. (The transferred pause adapter -reaches 66.5%, hard 38.9%, consistent with the same reading.) **Rust/MultiPL-E** (Python-trained, different +reaches 66.5%, hard 38.9%, consistent with the same reading.) + +**LiveCodeBench sharpens this into a dissociation** (150 newest stdin +problems, Nov 2024–Apr 2025, execution-verified; no LCB training anywhere +in the pipeline; base 18.7%): + +| arm (MBPP-trained where trained) | overall | hard (n=25) | vs base, paired | +|---|---|---|---| +| **untrained merge, k=4** | **24.0%** | **36.0%** | **+**, p=0.039 | +| trained loop, k=4 | 15.3% | 8.0% | −, p=0.23 | +| distill FF, k=1 | 12.7% | 16.0% | **−**, p=0.049 | + +Far from distribution, the *trained content is a liability* (distill +significantly hurts; untrained-vs-trained-loop is 14–1 discordant, +p=0.001) while the *untrained anchored recurrence significantly helps* — +the training-free regime of Lys et al. is the right choice off-distribution, +and the amortized-content reading of §3.3 predicts exactly this: what the +adapter learned is MBPP-shaped plan content, valuable where plans look like +MBPP plans and harmful where they don't. Transfer ordering by distance: +HumanEval (near) — trained ≈ untrained; Rust (mid) — trained helps the hard +bucket; LCB (far) — untrained wins outright. Caveats: single seed per arm, +hard n=25, one benchmark at the far end. **Rust/MultiPL-E** (Python-trained, different language, compile-run-verified): hard 8.0%→24.0% (p=0.125 at n=25 — directionally consistent, underpowered). **Blocksworld** MBPP-transfer: hard 0→14.3% (task-trained: 43%). Content transfers where the substrate's @@ -316,9 +342,22 @@ removing the easy-item perturbation tax. k=0 is the exact base model by construction — the implant is removable at token granularity. **General-capability panel** (ARC-Challenge, WinoGrande, HellaSwag, MMLU; -length-normalized MC scoring with the loop applied to the context span) is -running on the Spark; results will quantify what k>0 does to off-task -abilities. [PENDING — fill on completion.] +800 items each, length-normalized MC likelihood via the chat template, loop +applied to the context span). The safety answer is clean — **k>0 does not +damage general abilities**: + +| arm | ARC-C | WinoGrande | HellaSwag | MMLU | +|---|---|---|---|---| +| base (k=0) | 36.0 | 55.9 | 52.3 | 30.1 | +| loop k=2 (MBPP adapter) | 36.1 | 55.3 | 49.6 | 31.3 | +| distill FF (MBPP) | 41.8 | 56.6 | 57.0 | 31.8 | + +The loop arm is flat within noise (largest move −2.6 on HellaSwag, +unpaired n=800). The distill adapter *nominally improves* every benchmark +(+5.8 ARC, +4.8 HellaSwag) — consistent with §3.7's finding that these +implants carry a generically useful perturbation component, though +MC-likelihood scoring and generation quality are different regimes (see +the LCB result below before reading this as free capability). ### 3.9 Negative results with content @@ -374,9 +413,15 @@ are small (n=55/38/25); within-ladder orderings are not individually significant, and only the pooled hard effect and the HumanEval overall gain survive multiple-comparison scrutiny. Bucket membership derives from greedy labeling runs (consensus-k0 robustness check moves numbers <2 points, but -both checks share the base model; an independent difficulty proxy is an -open external check). A third architecture family was not run; -LiveCodeBench (contamination-safe) was not run; rung-2 was not run at 12B. The easy-item perturbation tax persists wherever the +both checks share the base model; an independent 12B-relabeling proxy is +running). A third architecture family was not run; rung-2 was not run at +12B. LiveCodeBench: single seed per arm, hard n=25, stdin-judged problems +only, and its newest shard (Apr 2025) is *newer than MBPP by years* but +not provably past the base model's undisclosed training cutoff — we claim +recency, not proven non-contamination. The capability panel is +MC-likelihood, not generation; its "no damage" answer does not extend to +generation quality off-distribution (LCB shows trained arms *do* hurt +there). The easy-item perturbation tax persists wherever the gate's precision fails. MBPP/GSM8K likely overlap pretraining data; both arms share contamination, and memorized items land in the easy bucket, but bucket composition is contamination-sensitive. The capability panel diff --git a/results-loop/eval_code_tap23_rerun.json b/results-loop/eval_code_tap23_rerun.json new file mode 100644 index 0000000..a3a06c9 --- /dev/null +++ b/results-loop/eval_code_tap23_rerun.json @@ -0,0 +1,3036 @@ +{ + "tag": "tap23_rerun", + "ks": { + "0": { + "acc": 0.492, + "by_label": { + "easy": 0.9836065573770492, + "hard": 0.03571428571428571, + "drop": 0.02 + }, + "per_item": [ + { + "task_id": 11, + "ok": true + }, + { + "task_id": 12, + "ok": false + }, + { + "task_id": 13, + "ok": false + }, + { + "task_id": 14, + "ok": true + }, + { + "task_id": 15, + "ok": false + }, + { + "task_id": 16, + "ok": false + }, + { + "task_id": 17, + "ok": true + }, + { + "task_id": 18, + "ok": true + }, + { + "task_id": 19, + "ok": true + }, + { + "task_id": 20, + "ok": false + }, + { + "task_id": 21, + "ok": false + }, + { + "task_id": 22, + "ok": true + }, + { + "task_id": 23, + "ok": true + }, + { + "task_id": 24, + "ok": false + }, + { + "task_id": 25, + "ok": true + }, + { + "task_id": 26, + "ok": false + }, + { + "task_id": 27, + "ok": true + }, + { + "task_id": 28, + "ok": true + }, + { + "task_id": 29, + "ok": true + }, + { + "task_id": 30, + "ok": true + }, + { + "task_id": 31, + "ok": false + }, + { + "task_id": 32, + "ok": true + }, + { + "task_id": 33, + "ok": false + }, + { + "task_id": 34, + "ok": false + }, + { + "task_id": 35, + "ok": false + }, + { + "task_id": 36, + "ok": false + }, + { + "task_id": 37, + "ok": true + }, + { + "task_id": 38, + "ok": false + }, + { + "task_id": 39, + "ok": false + }, + { + "task_id": 40, + "ok": true + }, + { + "task_id": 41, + "ok": true + }, + { + "task_id": 42, + "ok": false + }, + { + "task_id": 43, + "ok": false + }, + { + "task_id": 44, + "ok": true + }, + { + "task_id": 45, + "ok": true + }, + { + "task_id": 46, + "ok": true + }, + { + "task_id": 47, + "ok": false + }, + { + "task_id": 48, + "ok": false + }, + { + "task_id": 49, + "ok": true + }, + { + "task_id": 50, + "ok": true + }, + { + "task_id": 51, + "ok": true + }, + { + "task_id": 52, + "ok": true + }, + { + "task_id": 53, + "ok": true + }, + { + "task_id": 54, + "ok": false + }, + { + "task_id": 55, + "ok": true + }, + { + "task_id": 56, + "ok": true + }, + { + "task_id": 57, + "ok": true + }, + { + "task_id": 58, + "ok": true + }, + { + "task_id": 59, + "ok": false + }, + { + "task_id": 60, + "ok": false + }, + { + "task_id": 61, + "ok": true + }, + { + "task_id": 62, + "ok": true + }, + { + "task_id": 63, + "ok": true + }, + { + "task_id": 64, + "ok": true + }, + { + "task_id": 65, + "ok": true + }, + { + "task_id": 66, + "ok": true + }, + { + "task_id": 67, + "ok": false + }, + { + "task_id": 68, + "ok": true + }, + { + "task_id": 69, + "ok": true + }, + { + "task_id": 70, + "ok": true + }, + { + "task_id": 71, + "ok": false + }, + { + "task_id": 72, + "ok": false + }, + { + "task_id": 73, + "ok": true + }, + { + "task_id": 74, + "ok": false + }, + { + "task_id": 75, + "ok": true + }, + { + "task_id": 76, + "ok": true + }, + { + "task_id": 77, + "ok": false + }, + { + "task_id": 78, + "ok": true + }, + { + "task_id": 79, + "ok": true + }, + { + "task_id": 80, + "ok": true + }, + { + "task_id": 81, + "ok": false + }, + { + "task_id": 82, + "ok": true + }, + { + "task_id": 83, + "ok": false + }, + { + "task_id": 84, + "ok": false + }, + { + "task_id": 85, + "ok": true + }, + { + "task_id": 86, + "ok": true + }, + { + "task_id": 87, + "ok": false + }, + { + "task_id": 88, + "ok": true + }, + { + "task_id": 89, + "ok": true + }, + { + "task_id": 90, + "ok": true + }, + { + "task_id": 91, + "ok": true + }, + { + "task_id": 92, + "ok": false + }, + { + "task_id": 93, + "ok": true + }, + { + "task_id": 94, + "ok": true + }, + { + "task_id": 95, + "ok": false + }, + { + "task_id": 96, + "ok": true + }, + { + "task_id": 97, + "ok": true + }, + { + "task_id": 98, + "ok": true + }, + { + "task_id": 99, + "ok": true + }, + { + "task_id": 100, + "ok": false + }, + { + "task_id": 101, + "ok": false + }, + { + "task_id": 102, + "ok": true + }, + { + "task_id": 103, + "ok": false + }, + { + "task_id": 104, + "ok": true + }, + { + "task_id": 105, + "ok": true + }, + { + "task_id": 106, + "ok": false + }, + { + "task_id": 107, + "ok": false + }, + { + "task_id": 108, + "ok": false + }, + { + "task_id": 109, + "ok": false + }, + { + "task_id": 110, + "ok": false + }, + { + "task_id": 111, + "ok": false + }, + { + "task_id": 112, + "ok": false + }, + { + "task_id": 113, + "ok": true + }, + { + "task_id": 114, + "ok": false + }, + { + "task_id": 115, + "ok": true + }, + { + "task_id": 116, + "ok": true + }, + { + "task_id": 117, + "ok": true + }, + { + "task_id": 118, + "ok": false + }, + { + "task_id": 119, + "ok": false + }, + { + "task_id": 120, + "ok": false + }, + { + "task_id": 121, + "ok": false + }, + { + "task_id": 122, + "ok": false + }, + { + "task_id": 123, + "ok": true + }, + { + "task_id": 124, + "ok": false + }, + { + "task_id": 125, + "ok": false + }, + { + "task_id": 126, + "ok": false + }, + { + "task_id": 127, + "ok": true + }, + { + "task_id": 128, + "ok": true + }, + { + "task_id": 129, + "ok": false + }, + { + "task_id": 130, + "ok": false + }, + { + "task_id": 131, + "ok": false + }, + { + "task_id": 132, + "ok": true + }, + { + "task_id": 133, + "ok": true + }, + { + "task_id": 134, + "ok": false + }, + { + "task_id": 135, + "ok": true + }, + { + "task_id": 136, + "ok": false + }, + { + "task_id": 137, + "ok": false + }, + { + "task_id": 138, + "ok": false + }, + { + "task_id": 139, + "ok": false + }, + { + "task_id": 140, + "ok": false + }, + { + "task_id": 141, + "ok": false + }, + { + "task_id": 142, + "ok": false + }, + { + "task_id": 143, + "ok": false + }, + { + "task_id": 144, + "ok": false + }, + { + "task_id": 145, + "ok": true + }, + { + "task_id": 146, + "ok": false + }, + { + "task_id": 147, + "ok": false + }, + { + "task_id": 148, + "ok": false + }, + { + "task_id": 149, + "ok": false + }, + { + "task_id": 150, + "ok": false + }, + { + "task_id": 151, + "ok": true + }, + { + "task_id": 152, + "ok": false + }, + { + "task_id": 153, + "ok": true + }, + { + "task_id": 154, + "ok": true + }, + { + "task_id": 155, + "ok": false + }, + { + "task_id": 156, + "ok": true + }, + { + "task_id": 157, + "ok": true + }, + { + "task_id": 158, + "ok": false + }, + { + "task_id": 159, + "ok": false + }, + { + "task_id": 160, + "ok": false + }, + { + "task_id": 161, + "ok": true + }, + { + "task_id": 162, + "ok": true + }, + { + "task_id": 163, + "ok": true + }, + { + "task_id": 164, + "ok": false + }, + { + "task_id": 165, + "ok": false + }, + { + "task_id": 166, + "ok": false + }, + { + "task_id": 167, + "ok": true + }, + { + "task_id": 168, + "ok": true + }, + { + "task_id": 169, + "ok": false + }, + { + "task_id": 170, + "ok": false + }, + { + "task_id": 171, + "ok": true + }, + { + "task_id": 172, + "ok": true + }, + { + "task_id": 173, + "ok": true + }, + { + "task_id": 174, + "ok": true + }, + { + "task_id": 175, + "ok": true + }, + { + "task_id": 176, + "ok": true + }, + { + "task_id": 177, + "ok": false + }, + { + "task_id": 178, + "ok": true + }, + { + "task_id": 179, + "ok": true + }, + { + "task_id": 180, + "ok": false + }, + { + "task_id": 181, + "ok": false + }, + { + "task_id": 182, + "ok": false + }, + { + "task_id": 183, + "ok": false + }, + { + "task_id": 184, + "ok": false + }, + { + "task_id": 185, + "ok": false + }, + { + "task_id": 186, + "ok": true + }, + { + "task_id": 187, + "ok": false + }, + { + "task_id": 188, + "ok": false + }, + { + "task_id": 189, + "ok": false + }, + { + "task_id": 190, + "ok": false + }, + { + "task_id": 191, + "ok": true + }, + { + "task_id": 192, + "ok": true + }, + { + "task_id": 193, + "ok": false + }, + { + "task_id": 194, + "ok": true + }, + { + "task_id": 195, + "ok": false + }, + { + "task_id": 196, + "ok": true + }, + { + "task_id": 197, + "ok": true + }, + { + "task_id": 198, + "ok": false + }, + { + "task_id": 199, + "ok": true + }, + { + "task_id": 200, + "ok": true + }, + { + "task_id": 201, + "ok": true + }, + { + "task_id": 202, + "ok": true + }, + { + "task_id": 203, + "ok": true + }, + { + "task_id": 204, + "ok": true + }, + { + "task_id": 205, + "ok": false + }, + { + "task_id": 206, + "ok": false + }, + { + "task_id": 207, + "ok": false + }, + { + "task_id": 208, + "ok": false + }, + { + "task_id": 209, + "ok": false + }, + { + "task_id": 210, + "ok": true + }, + { + "task_id": 211, + "ok": false + }, + { + "task_id": 212, + "ok": false + }, + { + "task_id": 213, + "ok": true + }, + { + "task_id": 214, + "ok": true + }, + { + "task_id": 215, + "ok": false + }, + { + "task_id": 216, + "ok": false + }, + { + "task_id": 217, + "ok": true + }, + { + "task_id": 218, + "ok": false + }, + { + "task_id": 219, + "ok": false + }, + { + "task_id": 220, + "ok": false + }, + { + "task_id": 221, + "ok": true + }, + { + "task_id": 222, + "ok": true + }, + { + "task_id": 223, + "ok": false + }, + { + "task_id": 224, + "ok": true + }, + { + "task_id": 225, + "ok": false + }, + { + "task_id": 226, + "ok": true + }, + { + "task_id": 227, + "ok": true + }, + { + "task_id": 228, + "ok": false + }, + { + "task_id": 229, + "ok": false + }, + { + "task_id": 230, + "ok": true + }, + { + "task_id": 231, + "ok": false + }, + { + "task_id": 232, + "ok": true + }, + { + "task_id": 233, + "ok": false + }, + { + "task_id": 234, + "ok": true + }, + { + "task_id": 235, + "ok": false + }, + { + "task_id": 236, + "ok": false + }, + { + "task_id": 237, + "ok": false + }, + { + "task_id": 238, + "ok": true + }, + { + "task_id": 239, + "ok": false + }, + { + "task_id": 240, + "ok": true + }, + { + "task_id": 241, + "ok": false + }, + { + "task_id": 242, + "ok": true + }, + { + "task_id": 243, + "ok": false + }, + { + "task_id": 244, + "ok": true + }, + { + "task_id": 245, + "ok": false + }, + { + "task_id": 246, + "ok": true + }, + { + "task_id": 247, + "ok": false + }, + { + "task_id": 248, + "ok": false + }, + { + "task_id": 249, + "ok": false + }, + { + "task_id": 250, + "ok": true + }, + { + "task_id": 251, + "ok": true + }, + { + "task_id": 252, + "ok": true + }, + { + "task_id": 253, + "ok": true + }, + { + "task_id": 254, + "ok": true + }, + { + "task_id": 255, + "ok": false + }, + { + "task_id": 256, + "ok": false + }, + { + "task_id": 257, + "ok": true + }, + { + "task_id": 258, + "ok": false + }, + { + "task_id": 259, + "ok": false + }, + { + "task_id": 260, + "ok": false + } + ] + }, + "2": { + "acc": 0.532, + "by_label": { + "easy": 0.9098360655737705, + "hard": 0.4642857142857143, + "drop": 0.09 + }, + "per_item": [ + { + "task_id": 11, + "ok": true + }, + { + "task_id": 12, + "ok": true + }, + { + "task_id": 13, + "ok": false + }, + { + "task_id": 14, + "ok": true + }, + { + "task_id": 15, + "ok": false + }, + { + "task_id": 16, + "ok": true + }, + { + "task_id": 17, + "ok": true + }, + { + "task_id": 18, + "ok": true + }, + { + "task_id": 19, + "ok": true + }, + { + "task_id": 20, + "ok": false + }, + { + "task_id": 21, + "ok": false + }, + { + "task_id": 22, + "ok": true + }, + { + "task_id": 23, + "ok": true + }, + { + "task_id": 24, + "ok": false + }, + { + "task_id": 25, + "ok": true + }, + { + "task_id": 26, + "ok": false + }, + { + "task_id": 27, + "ok": true + }, + { + "task_id": 28, + "ok": true + }, + { + "task_id": 29, + "ok": true + }, + { + "task_id": 30, + "ok": true + }, + { + "task_id": 31, + "ok": false + }, + { + "task_id": 32, + "ok": true + }, + { + "task_id": 33, + "ok": false + }, + { + "task_id": 34, + "ok": false + }, + { + "task_id": 35, + "ok": false + }, + { + "task_id": 36, + "ok": false + }, + { + "task_id": 37, + "ok": true + }, + { + "task_id": 38, + "ok": false + }, + { + "task_id": 39, + "ok": false + }, + { + "task_id": 40, + "ok": true + }, + { + "task_id": 41, + "ok": true + }, + { + "task_id": 42, + "ok": false + }, + { + "task_id": 43, + "ok": false + }, + { + "task_id": 44, + "ok": false + }, + { + "task_id": 45, + "ok": true + }, + { + "task_id": 46, + "ok": true + }, + { + "task_id": 47, + "ok": false + }, + { + "task_id": 48, + "ok": false + }, + { + "task_id": 49, + "ok": true + }, + { + "task_id": 50, + "ok": true + }, + { + "task_id": 51, + "ok": true + }, + { + "task_id": 52, + "ok": true + }, + { + "task_id": 53, + "ok": true + }, + { + "task_id": 54, + "ok": true + }, + { + "task_id": 55, + "ok": true + }, + { + "task_id": 56, + "ok": true + }, + { + "task_id": 57, + "ok": false + }, + { + "task_id": 58, + "ok": true + }, + { + "task_id": 59, + "ok": false + }, + { + "task_id": 60, + "ok": false + }, + { + "task_id": 61, + "ok": true + }, + { + "task_id": 62, + "ok": true + }, + { + "task_id": 63, + "ok": true + }, + { + "task_id": 64, + "ok": true + }, + { + "task_id": 65, + "ok": true + }, + { + "task_id": 66, + "ok": true + }, + { + "task_id": 67, + "ok": false + }, + { + "task_id": 68, + "ok": true + }, + { + "task_id": 69, + "ok": true + }, + { + "task_id": 70, + "ok": true + }, + { + "task_id": 71, + "ok": false + }, + { + "task_id": 72, + "ok": true + }, + { + "task_id": 73, + "ok": false + }, + { + "task_id": 74, + "ok": false + }, + { + "task_id": 75, + "ok": true + }, + { + "task_id": 76, + "ok": true + }, + { + "task_id": 77, + "ok": false + }, + { + "task_id": 78, + "ok": true + }, + { + "task_id": 79, + "ok": true + }, + { + "task_id": 80, + "ok": true + }, + { + "task_id": 81, + "ok": false + }, + { + "task_id": 82, + "ok": true + }, + { + "task_id": 83, + "ok": false + }, + { + "task_id": 84, + "ok": false + }, + { + "task_id": 85, + "ok": true + }, + { + "task_id": 86, + "ok": true + }, + { + "task_id": 87, + "ok": false + }, + { + "task_id": 88, + "ok": true + }, + { + "task_id": 89, + "ok": true + }, + { + "task_id": 90, + "ok": true + }, + { + "task_id": 91, + "ok": true + }, + { + "task_id": 92, + "ok": false + }, + { + "task_id": 93, + "ok": true + }, + { + "task_id": 94, + "ok": true + }, + { + "task_id": 95, + "ok": true + }, + { + "task_id": 96, + "ok": true + }, + { + "task_id": 97, + "ok": true + }, + { + "task_id": 98, + "ok": true + }, + { + "task_id": 99, + "ok": true + }, + { + "task_id": 100, + "ok": false + }, + { + "task_id": 101, + "ok": false + }, + { + "task_id": 102, + "ok": true + }, + { + "task_id": 103, + "ok": false + }, + { + "task_id": 104, + "ok": true + }, + { + "task_id": 105, + "ok": false + }, + { + "task_id": 106, + "ok": false + }, + { + "task_id": 107, + "ok": false + }, + { + "task_id": 108, + "ok": false + }, + { + "task_id": 109, + "ok": true + }, + { + "task_id": 110, + "ok": false + }, + { + "task_id": 111, + "ok": false + }, + { + "task_id": 112, + "ok": false + }, + { + "task_id": 113, + "ok": false + }, + { + "task_id": 114, + "ok": false + }, + { + "task_id": 115, + "ok": true + }, + { + "task_id": 116, + "ok": true + }, + { + "task_id": 117, + "ok": true + }, + { + "task_id": 118, + "ok": true + }, + { + "task_id": 119, + "ok": false + }, + { + "task_id": 120, + "ok": true + }, + { + "task_id": 121, + "ok": false + }, + { + "task_id": 122, + "ok": false + }, + { + "task_id": 123, + "ok": false + }, + { + "task_id": 124, + "ok": false + }, + { + "task_id": 125, + "ok": false + }, + { + "task_id": 126, + "ok": false + }, + { + "task_id": 127, + "ok": true + }, + { + "task_id": 128, + "ok": true + }, + { + "task_id": 129, + "ok": false + }, + { + "task_id": 130, + "ok": true + }, + { + "task_id": 131, + "ok": true + }, + { + "task_id": 132, + "ok": true + }, + { + "task_id": 133, + "ok": true + }, + { + "task_id": 134, + "ok": false + }, + { + "task_id": 135, + "ok": true + }, + { + "task_id": 136, + "ok": false + }, + { + "task_id": 137, + "ok": false + }, + { + "task_id": 138, + "ok": false + }, + { + "task_id": 139, + "ok": false + }, + { + "task_id": 140, + "ok": false + }, + { + "task_id": 141, + "ok": false + }, + { + "task_id": 142, + "ok": false + }, + { + "task_id": 143, + "ok": false + }, + { + "task_id": 144, + "ok": false + }, + { + "task_id": 145, + "ok": true + }, + { + "task_id": 146, + "ok": false + }, + { + "task_id": 147, + "ok": false + }, + { + "task_id": 148, + "ok": false + }, + { + "task_id": 149, + "ok": true + }, + { + "task_id": 150, + "ok": false + }, + { + "task_id": 151, + "ok": true + }, + { + "task_id": 152, + "ok": false + }, + { + "task_id": 153, + "ok": true + }, + { + "task_id": 154, + "ok": true + }, + { + "task_id": 155, + "ok": false + }, + { + "task_id": 156, + "ok": true + }, + { + "task_id": 157, + "ok": true + }, + { + "task_id": 158, + "ok": false + }, + { + "task_id": 159, + "ok": true + }, + { + "task_id": 160, + "ok": false + }, + { + "task_id": 161, + "ok": true + }, + { + "task_id": 162, + "ok": true + }, + { + "task_id": 163, + "ok": true + }, + { + "task_id": 164, + "ok": false + }, + { + "task_id": 165, + "ok": true + }, + { + "task_id": 166, + "ok": true + }, + { + "task_id": 167, + "ok": true + }, + { + "task_id": 168, + "ok": true + }, + { + "task_id": 169, + "ok": false + }, + { + "task_id": 170, + "ok": true + }, + { + "task_id": 171, + "ok": true + }, + { + "task_id": 172, + "ok": false + }, + { + "task_id": 173, + "ok": true + }, + { + "task_id": 174, + "ok": true + }, + { + "task_id": 175, + "ok": true + }, + { + "task_id": 176, + "ok": true + }, + { + "task_id": 177, + "ok": false + }, + { + "task_id": 178, + "ok": true + }, + { + "task_id": 179, + "ok": true + }, + { + "task_id": 180, + "ok": false + }, + { + "task_id": 181, + "ok": false + }, + { + "task_id": 182, + "ok": false + }, + { + "task_id": 183, + "ok": false + }, + { + "task_id": 184, + "ok": false + }, + { + "task_id": 185, + "ok": false + }, + { + "task_id": 186, + "ok": true + }, + { + "task_id": 187, + "ok": true + }, + { + "task_id": 188, + "ok": false + }, + { + "task_id": 189, + "ok": false + }, + { + "task_id": 190, + "ok": false + }, + { + "task_id": 191, + "ok": true + }, + { + "task_id": 192, + "ok": true + }, + { + "task_id": 193, + "ok": true + }, + { + "task_id": 194, + "ok": false + }, + { + "task_id": 195, + "ok": true + }, + { + "task_id": 196, + "ok": true + }, + { + "task_id": 197, + "ok": true + }, + { + "task_id": 198, + "ok": false + }, + { + "task_id": 199, + "ok": true + }, + { + "task_id": 200, + "ok": true + }, + { + "task_id": 201, + "ok": true + }, + { + "task_id": 202, + "ok": true + }, + { + "task_id": 203, + "ok": true + }, + { + "task_id": 204, + "ok": true + }, + { + "task_id": 205, + "ok": false + }, + { + "task_id": 206, + "ok": false + }, + { + "task_id": 207, + "ok": false + }, + { + "task_id": 208, + "ok": false + }, + { + "task_id": 209, + "ok": false + }, + { + "task_id": 210, + "ok": true + }, + { + "task_id": 211, + "ok": false + }, + { + "task_id": 212, + "ok": true + }, + { + "task_id": 213, + "ok": false + }, + { + "task_id": 214, + "ok": true + }, + { + "task_id": 215, + "ok": false + }, + { + "task_id": 216, + "ok": false + }, + { + "task_id": 217, + "ok": false + }, + { + "task_id": 218, + "ok": false + }, + { + "task_id": 219, + "ok": false + }, + { + "task_id": 220, + "ok": false + }, + { + "task_id": 221, + "ok": true + }, + { + "task_id": 222, + "ok": true + }, + { + "task_id": 223, + "ok": false + }, + { + "task_id": 224, + "ok": true + }, + { + "task_id": 225, + "ok": true + }, + { + "task_id": 226, + "ok": true + }, + { + "task_id": 227, + "ok": true + }, + { + "task_id": 228, + "ok": false + }, + { + "task_id": 229, + "ok": false + }, + { + "task_id": 230, + "ok": false + }, + { + "task_id": 231, + "ok": true + }, + { + "task_id": 232, + "ok": true + }, + { + "task_id": 233, + "ok": false + }, + { + "task_id": 234, + "ok": true + }, + { + "task_id": 235, + "ok": false + }, + { + "task_id": 236, + "ok": false + }, + { + "task_id": 237, + "ok": false + }, + { + "task_id": 238, + "ok": true + }, + { + "task_id": 239, + "ok": false + }, + { + "task_id": 240, + "ok": true + }, + { + "task_id": 241, + "ok": false + }, + { + "task_id": 242, + "ok": true + }, + { + "task_id": 243, + "ok": false + }, + { + "task_id": 244, + "ok": true + }, + { + "task_id": 245, + "ok": false + }, + { + "task_id": 246, + "ok": false + }, + { + "task_id": 247, + "ok": false + }, + { + "task_id": 248, + "ok": false + }, + { + "task_id": 249, + "ok": true + }, + { + "task_id": 250, + "ok": true + }, + { + "task_id": 251, + "ok": true + }, + { + "task_id": 252, + "ok": true + }, + { + "task_id": 253, + "ok": true + }, + { + "task_id": 254, + "ok": false + }, + { + "task_id": 255, + "ok": false + }, + { + "task_id": 256, + "ok": false + }, + { + "task_id": 257, + "ok": true + }, + { + "task_id": 258, + "ok": true + }, + { + "task_id": 259, + "ok": false + }, + { + "task_id": 260, + "ok": false + } + ] + }, + "4": { + "acc": 0.524, + "by_label": { + "easy": 0.8934426229508197, + "hard": 0.42857142857142855, + "drop": 0.1 + }, + "per_item": [ + { + "task_id": 11, + "ok": true + }, + { + "task_id": 12, + "ok": true + }, + { + "task_id": 13, + "ok": false + }, + { + "task_id": 14, + "ok": true + }, + { + "task_id": 15, + "ok": false + }, + { + "task_id": 16, + "ok": false + }, + { + "task_id": 17, + "ok": true + }, + { + "task_id": 18, + "ok": true + }, + { + "task_id": 19, + "ok": true + }, + { + "task_id": 20, + "ok": false + }, + { + "task_id": 21, + "ok": false + }, + { + "task_id": 22, + "ok": true + }, + { + "task_id": 23, + "ok": true + }, + { + "task_id": 24, + "ok": false + }, + { + "task_id": 25, + "ok": true + }, + { + "task_id": 26, + "ok": false + }, + { + "task_id": 27, + "ok": true + }, + { + "task_id": 28, + "ok": true + }, + { + "task_id": 29, + "ok": true + }, + { + "task_id": 30, + "ok": true + }, + { + "task_id": 31, + "ok": false + }, + { + "task_id": 32, + "ok": true + }, + { + "task_id": 33, + "ok": false + }, + { + "task_id": 34, + "ok": false + }, + { + "task_id": 35, + "ok": true + }, + { + "task_id": 36, + "ok": false + }, + { + "task_id": 37, + "ok": true + }, + { + "task_id": 38, + "ok": false + }, + { + "task_id": 39, + "ok": false + }, + { + "task_id": 40, + "ok": true + }, + { + "task_id": 41, + "ok": true + }, + { + "task_id": 42, + "ok": false + }, + { + "task_id": 43, + "ok": false + }, + { + "task_id": 44, + "ok": false + }, + { + "task_id": 45, + "ok": true + }, + { + "task_id": 46, + "ok": true + }, + { + "task_id": 47, + "ok": false + }, + { + "task_id": 48, + "ok": false + }, + { + "task_id": 49, + "ok": true + }, + { + "task_id": 50, + "ok": false + }, + { + "task_id": 51, + "ok": true + }, + { + "task_id": 52, + "ok": true + }, + { + "task_id": 53, + "ok": true + }, + { + "task_id": 54, + "ok": true + }, + { + "task_id": 55, + "ok": true + }, + { + "task_id": 56, + "ok": true + }, + { + "task_id": 57, + "ok": true + }, + { + "task_id": 58, + "ok": true + }, + { + "task_id": 59, + "ok": false + }, + { + "task_id": 60, + "ok": true + }, + { + "task_id": 61, + "ok": true + }, + { + "task_id": 62, + "ok": true + }, + { + "task_id": 63, + "ok": false + }, + { + "task_id": 64, + "ok": true + }, + { + "task_id": 65, + "ok": true + }, + { + "task_id": 66, + "ok": true + }, + { + "task_id": 67, + "ok": false + }, + { + "task_id": 68, + "ok": true + }, + { + "task_id": 69, + "ok": true + }, + { + "task_id": 70, + "ok": true + }, + { + "task_id": 71, + "ok": false + }, + { + "task_id": 72, + "ok": false + }, + { + "task_id": 73, + "ok": false + }, + { + "task_id": 74, + "ok": false + }, + { + "task_id": 75, + "ok": true + }, + { + "task_id": 76, + "ok": true + }, + { + "task_id": 77, + "ok": false + }, + { + "task_id": 78, + "ok": true + }, + { + "task_id": 79, + "ok": true + }, + { + "task_id": 80, + "ok": true + }, + { + "task_id": 81, + "ok": false + }, + { + "task_id": 82, + "ok": true + }, + { + "task_id": 83, + "ok": false + }, + { + "task_id": 84, + "ok": false + }, + { + "task_id": 85, + "ok": true + }, + { + "task_id": 86, + "ok": true + }, + { + "task_id": 87, + "ok": false + }, + { + "task_id": 88, + "ok": true + }, + { + "task_id": 89, + "ok": false + }, + { + "task_id": 90, + "ok": true + }, + { + "task_id": 91, + "ok": true + }, + { + "task_id": 92, + "ok": false + }, + { + "task_id": 93, + "ok": true + }, + { + "task_id": 94, + "ok": true + }, + { + "task_id": 95, + "ok": true + }, + { + "task_id": 96, + "ok": true + }, + { + "task_id": 97, + "ok": true + }, + { + "task_id": 98, + "ok": true + }, + { + "task_id": 99, + "ok": true + }, + { + "task_id": 100, + "ok": false + }, + { + "task_id": 101, + "ok": false + }, + { + "task_id": 102, + "ok": true + }, + { + "task_id": 103, + "ok": false + }, + { + "task_id": 104, + "ok": true + }, + { + "task_id": 105, + "ok": true + }, + { + "task_id": 106, + "ok": false + }, + { + "task_id": 107, + "ok": false + }, + { + "task_id": 108, + "ok": false + }, + { + "task_id": 109, + "ok": false + }, + { + "task_id": 110, + "ok": false + }, + { + "task_id": 111, + "ok": false + }, + { + "task_id": 112, + "ok": false + }, + { + "task_id": 113, + "ok": true + }, + { + "task_id": 114, + "ok": false + }, + { + "task_id": 115, + "ok": true + }, + { + "task_id": 116, + "ok": true + }, + { + "task_id": 117, + "ok": false + }, + { + "task_id": 118, + "ok": true + }, + { + "task_id": 119, + "ok": false + }, + { + "task_id": 120, + "ok": true + }, + { + "task_id": 121, + "ok": false + }, + { + "task_id": 122, + "ok": false + }, + { + "task_id": 123, + "ok": false + }, + { + "task_id": 124, + "ok": false + }, + { + "task_id": 125, + "ok": false + }, + { + "task_id": 126, + "ok": false + }, + { + "task_id": 127, + "ok": true + }, + { + "task_id": 128, + "ok": true + }, + { + "task_id": 129, + "ok": false + }, + { + "task_id": 130, + "ok": true + }, + { + "task_id": 131, + "ok": true + }, + { + "task_id": 132, + "ok": true + }, + { + "task_id": 133, + "ok": true + }, + { + "task_id": 134, + "ok": false + }, + { + "task_id": 135, + "ok": true + }, + { + "task_id": 136, + "ok": false + }, + { + "task_id": 137, + "ok": false + }, + { + "task_id": 138, + "ok": false + }, + { + "task_id": 139, + "ok": false + }, + { + "task_id": 140, + "ok": false + }, + { + "task_id": 141, + "ok": false + }, + { + "task_id": 142, + "ok": false + }, + { + "task_id": 143, + "ok": false + }, + { + "task_id": 144, + "ok": true + }, + { + "task_id": 145, + "ok": true + }, + { + "task_id": 146, + "ok": false + }, + { + "task_id": 147, + "ok": false + }, + { + "task_id": 148, + "ok": false + }, + { + "task_id": 149, + "ok": true + }, + { + "task_id": 150, + "ok": false + }, + { + "task_id": 151, + "ok": true + }, + { + "task_id": 152, + "ok": false + }, + { + "task_id": 153, + "ok": false + }, + { + "task_id": 154, + "ok": true + }, + { + "task_id": 155, + "ok": false + }, + { + "task_id": 156, + "ok": true + }, + { + "task_id": 157, + "ok": true + }, + { + "task_id": 158, + "ok": false + }, + { + "task_id": 159, + "ok": true + }, + { + "task_id": 160, + "ok": false + }, + { + "task_id": 161, + "ok": true + }, + { + "task_id": 162, + "ok": true + }, + { + "task_id": 163, + "ok": true + }, + { + "task_id": 164, + "ok": false + }, + { + "task_id": 165, + "ok": true + }, + { + "task_id": 166, + "ok": true + }, + { + "task_id": 167, + "ok": true + }, + { + "task_id": 168, + "ok": true + }, + { + "task_id": 169, + "ok": false + }, + { + "task_id": 170, + "ok": true + }, + { + "task_id": 171, + "ok": true + }, + { + "task_id": 172, + "ok": true + }, + { + "task_id": 173, + "ok": true + }, + { + "task_id": 174, + "ok": true + }, + { + "task_id": 175, + "ok": false + }, + { + "task_id": 176, + "ok": true + }, + { + "task_id": 177, + "ok": false + }, + { + "task_id": 178, + "ok": true + }, + { + "task_id": 179, + "ok": false + }, + { + "task_id": 180, + "ok": false + }, + { + "task_id": 181, + "ok": false + }, + { + "task_id": 182, + "ok": false + }, + { + "task_id": 183, + "ok": false + }, + { + "task_id": 184, + "ok": false + }, + { + "task_id": 185, + "ok": false + }, + { + "task_id": 186, + "ok": true + }, + { + "task_id": 187, + "ok": true + }, + { + "task_id": 188, + "ok": false + }, + { + "task_id": 189, + "ok": false + }, + { + "task_id": 190, + "ok": false + }, + { + "task_id": 191, + "ok": true + }, + { + "task_id": 192, + "ok": true + }, + { + "task_id": 193, + "ok": true + }, + { + "task_id": 194, + "ok": false + }, + { + "task_id": 195, + "ok": true + }, + { + "task_id": 196, + "ok": true + }, + { + "task_id": 197, + "ok": true + }, + { + "task_id": 198, + "ok": false + }, + { + "task_id": 199, + "ok": true + }, + { + "task_id": 200, + "ok": true + }, + { + "task_id": 201, + "ok": true + }, + { + "task_id": 202, + "ok": true + }, + { + "task_id": 203, + "ok": true + }, + { + "task_id": 204, + "ok": true + }, + { + "task_id": 205, + "ok": false + }, + { + "task_id": 206, + "ok": false + }, + { + "task_id": 207, + "ok": false + }, + { + "task_id": 208, + "ok": true + }, + { + "task_id": 209, + "ok": false + }, + { + "task_id": 210, + "ok": true + }, + { + "task_id": 211, + "ok": false + }, + { + "task_id": 212, + "ok": true + }, + { + "task_id": 213, + "ok": false + }, + { + "task_id": 214, + "ok": true + }, + { + "task_id": 215, + "ok": false + }, + { + "task_id": 216, + "ok": false + }, + { + "task_id": 217, + "ok": false + }, + { + "task_id": 218, + "ok": false + }, + { + "task_id": 219, + "ok": false + }, + { + "task_id": 220, + "ok": false + }, + { + "task_id": 221, + "ok": true + }, + { + "task_id": 222, + "ok": true + }, + { + "task_id": 223, + "ok": false + }, + { + "task_id": 224, + "ok": true + }, + { + "task_id": 225, + "ok": true + }, + { + "task_id": 226, + "ok": true + }, + { + "task_id": 227, + "ok": true + }, + { + "task_id": 228, + "ok": false + }, + { + "task_id": 229, + "ok": false + }, + { + "task_id": 230, + "ok": true + }, + { + "task_id": 231, + "ok": false + }, + { + "task_id": 232, + "ok": true + }, + { + "task_id": 233, + "ok": false + }, + { + "task_id": 234, + "ok": true + }, + { + "task_id": 235, + "ok": false + }, + { + "task_id": 236, + "ok": false + }, + { + "task_id": 237, + "ok": false + }, + { + "task_id": 238, + "ok": true + }, + { + "task_id": 239, + "ok": false + }, + { + "task_id": 240, + "ok": false + }, + { + "task_id": 241, + "ok": false + }, + { + "task_id": 242, + "ok": true + }, + { + "task_id": 243, + "ok": false + }, + { + "task_id": 244, + "ok": true + }, + { + "task_id": 245, + "ok": false + }, + { + "task_id": 246, + "ok": false + }, + { + "task_id": 247, + "ok": false + }, + { + "task_id": 248, + "ok": false + }, + { + "task_id": 249, + "ok": true + }, + { + "task_id": 250, + "ok": true + }, + { + "task_id": 251, + "ok": true + }, + { + "task_id": 252, + "ok": true + }, + { + "task_id": 253, + "ok": true + }, + { + "task_id": 254, + "ok": false + }, + { + "task_id": 255, + "ok": false + }, + { + "task_id": 256, + "ok": true + }, + { + "task_id": 257, + "ok": true + }, + { + "task_id": 258, + "ok": true + }, + { + "task_id": 259, + "ok": false + }, + { + "task_id": 260, + "ok": false + } + ] + } + }, + "n": 250 +} \ No newline at end of file diff --git a/results-loop/eval_lcb_distill.json b/results-loop/eval_lcb_distill.json new file mode 100644 index 0000000..bbc7ff7 --- /dev/null +++ b/results-loop/eval_lcb_distill.json @@ -0,0 +1,1221 @@ +{ + "tag": "distill", + "since": "2024-07-01", + "n": 150, + "ks": { + "0": { + "acc": 0.18666666666666668, + "hard_n": 25, + "hard_acc": 0.0, + "per_item": [ + { + "task_id": "abc382_a", + "ok": true + }, + { + "task_id": "abc382_b", + "ok": true + }, + { + "task_id": "abc382_c", + "ok": true + }, + { + "task_id": "abc382_d", + "ok": false + }, + { + "task_id": "abc382_f", + "ok": false + }, + { + "task_id": "abc382_g", + "ok": false + }, + { + "task_id": "abc383_a", + "ok": true + }, + { + "task_id": "abc383_b", + "ok": true + }, + { + "task_id": "abc383_c", + "ok": false + }, + { + "task_id": "abc383_d", + "ok": false + }, + { + "task_id": "abc383_e", + "ok": false + }, + { + "task_id": "abc384_a", + "ok": true + }, + { + "task_id": "abc384_b", + "ok": true + }, + { + "task_id": "abc384_c", + "ok": false + }, + { + "task_id": "abc384_d", + "ok": false + }, + { + "task_id": "abc384_e", + "ok": false + }, + { + "task_id": "abc384_f", + "ok": false + }, + { + "task_id": "abc384_g", + "ok": true + }, + { + "task_id": "abc385_a", + "ok": false + }, + { + "task_id": "abc385_b", + "ok": false + }, + { + "task_id": "abc385_c", + "ok": false + }, + { + "task_id": "abc385_d", + "ok": false + }, + { + "task_id": "abc385_e", + "ok": false + }, + { + "task_id": "abc385_f", + "ok": false + }, + { + "task_id": "abc386_a", + "ok": false + }, + { + "task_id": "abc386_b", + "ok": false + }, + { + "task_id": "abc386_c", + "ok": false + }, + { + "task_id": "abc386_d", + "ok": false + }, + { + "task_id": "abc386_e", + "ok": false + }, + { + "task_id": "abc386_f", + "ok": false + }, + { + "task_id": "abc387_a", + "ok": true + }, + { + "task_id": "abc387_b", + "ok": true + }, + { + "task_id": "abc387_c", + "ok": false + }, + { + "task_id": "abc387_f", + "ok": false + }, + { + "task_id": "abc388_a", + "ok": true + }, + { + "task_id": "abc388_b", + "ok": true + }, + { + "task_id": "abc388_c", + "ok": false + }, + { + "task_id": "abc388_d", + "ok": false + }, + { + "task_id": "abc388_e", + "ok": false + }, + { + "task_id": "abc388_f", + "ok": false + }, + { + "task_id": "abc388_g", + "ok": false + }, + { + "task_id": "abc389_a", + "ok": true + }, + { + "task_id": "abc389_b", + "ok": false + }, + { + "task_id": "abc389_d", + "ok": false + }, + { + "task_id": "abc389_e", + "ok": false + }, + { + "task_id": "abc389_f", + "ok": false + }, + { + "task_id": "abc389_g", + "ok": false + }, + { + "task_id": "abc390_a", + "ok": true + }, + { + "task_id": "abc390_b", + "ok": false + }, + { + "task_id": "abc390_c", + "ok": false + }, + { + "task_id": "abc390_d", + "ok": false + }, + { + "task_id": "abc390_e", + "ok": false + }, + { + "task_id": "abc390_f", + "ok": false + }, + { + "task_id": "abc390_g", + "ok": false + }, + { + "task_id": "abc391_a", + "ok": true + }, + { + "task_id": "abc391_b", + "ok": true + }, + { + "task_id": "abc391_d", + "ok": false + }, + { + "task_id": "abc391_e", + "ok": false + }, + { + "task_id": "abc391_f", + "ok": false + }, + { + "task_id": "abc391_g", + "ok": false + }, + { + "task_id": "abc392_a", + "ok": true + }, + { + "task_id": "abc392_b", + "ok": true + }, + { + "task_id": "abc392_c", + "ok": false + }, + { + "task_id": "abc392_d", + "ok": false + }, + { + "task_id": "abc392_f", + "ok": false + }, + { + "task_id": "abc392_g", + "ok": false + }, + { + "task_id": "abc393_a", + "ok": false + }, + { + "task_id": "abc393_b", + "ok": true + }, + { + "task_id": "abc393_d", + "ok": false + }, + { + "task_id": "abc393_e", + "ok": false + }, + { + "task_id": "abc393_f", + "ok": false + }, + { + "task_id": "abc394_a", + "ok": true + }, + { + "task_id": "abc394_b", + "ok": true + }, + { + "task_id": "abc394_c", + "ok": false + }, + { + "task_id": "abc394_d", + "ok": false + }, + { + "task_id": "abc394_e", + "ok": false + }, + { + "task_id": "abc394_f", + "ok": false + }, + { + "task_id": "abc394_g", + "ok": false + }, + { + "task_id": "abc395_a", + "ok": true + }, + { + "task_id": "abc395_b", + "ok": false + }, + { + "task_id": "abc395_c", + "ok": false + }, + { + "task_id": "abc395_e", + "ok": false + }, + { + "task_id": "abc395_f", + "ok": false + }, + { + "task_id": "abc396_a", + "ok": true + }, + { + "task_id": "abc396_b", + "ok": true + }, + { + "task_id": "abc396_c", + "ok": false + }, + { + "task_id": "abc396_d", + "ok": false + }, + { + "task_id": "abc396_e", + "ok": false + }, + { + "task_id": "abc396_f", + "ok": false + }, + { + "task_id": "abc396_g", + "ok": false + }, + { + "task_id": "abc397_a", + "ok": true + }, + { + "task_id": "abc397_b", + "ok": false + }, + { + "task_id": "abc397_c", + "ok": false + }, + { + "task_id": "abc397_d", + "ok": false + }, + { + "task_id": "abc397_e", + "ok": false + }, + { + "task_id": "abc397_f", + "ok": false + }, + { + "task_id": "abc397_g", + "ok": false + }, + { + "task_id": "abc398_a", + "ok": false + }, + { + "task_id": "abc398_b", + "ok": true + }, + { + "task_id": "abc398_c", + "ok": false + }, + { + "task_id": "abc398_d", + "ok": false + }, + { + "task_id": "abc398_f", + "ok": false + }, + { + "task_id": "abc398_g", + "ok": false + }, + { + "task_id": "abc399_a", + "ok": true + }, + { + "task_id": "abc399_b", + "ok": false + }, + { + "task_id": "abc399_c", + "ok": false + }, + { + "task_id": "abc399_d", + "ok": false + }, + { + "task_id": "abc399_e", + "ok": false + }, + { + "task_id": "abc399_f", + "ok": false + }, + { + "task_id": "abc400_a", + "ok": true + }, + { + "task_id": "abc400_b", + "ok": false + }, + { + "task_id": "abc400_c", + "ok": false + }, + { + "task_id": "abc400_d", + "ok": false + }, + { + "task_id": "abc400_e", + "ok": false + }, + { + "task_id": "abc400_g", + "ok": false + }, + { + "task_id": "arc188_a", + "ok": false + }, + { + "task_id": "arc188_b", + "ok": false + }, + { + "task_id": "arc188_c", + "ok": false + }, + { + "task_id": "arc188_d", + "ok": false + }, + { + "task_id": "arc189_a", + "ok": false + }, + { + "task_id": "arc189_b", + "ok": false + }, + { + "task_id": "arc189_c", + "ok": false + }, + { + "task_id": "arc189_d", + "ok": false + }, + { + "task_id": "arc190_a", + "ok": false + }, + { + "task_id": "arc190_c", + "ok": false + }, + { + "task_id": "arc190_d", + "ok": false + }, + { + "task_id": "arc191_a", + "ok": false + }, + { + "task_id": "arc191_c", + "ok": false + }, + { + "task_id": "arc191_d", + "ok": false + }, + { + "task_id": "arc192_a", + "ok": false + }, + { + "task_id": "arc192_b", + "ok": false + }, + { + "task_id": "arc192_d", + "ok": false + }, + { + "task_id": "arc192_e", + "ok": false + }, + { + "task_id": "arc193_a", + "ok": false + }, + { + "task_id": "arc193_b", + "ok": false + }, + { + "task_id": "arc193_d", + "ok": false + }, + { + "task_id": "arc194_a", + "ok": false + }, + { + "task_id": "arc194_b", + "ok": false + }, + { + "task_id": "arc194_c", + "ok": false + }, + { + "task_id": "arc194_d", + "ok": false + }, + { + "task_id": "arc194_e", + "ok": false + }, + { + "task_id": "arc195_a", + "ok": false + }, + { + "task_id": "arc195_b", + "ok": false + }, + { + "task_id": "arc195_c", + "ok": false + }, + { + "task_id": "arc195_d", + "ok": false + }, + { + "task_id": "arc195_e", + "ok": false + }, + { + "task_id": "arc196_a", + "ok": false + }, + { + "task_id": "arc196_b", + "ok": false + }, + { + "task_id": "arc196_c", + "ok": false + }, + { + "task_id": "arc196_d", + "ok": false + } + ] + }, + "1": { + "acc": 0.12666666666666668, + "hard_n": 25, + "hard_acc": 0.16, + "per_item": [ + { + "task_id": "abc382_a", + "ok": true + }, + { + "task_id": "abc382_b", + "ok": false + }, + { + "task_id": "abc382_c", + "ok": false + }, + { + "task_id": "abc382_d", + "ok": true + }, + { + "task_id": "abc382_f", + "ok": false + }, + { + "task_id": "abc382_g", + "ok": false + }, + { + "task_id": "abc383_a", + "ok": false + }, + { + "task_id": "abc383_b", + "ok": false + }, + { + "task_id": "abc383_c", + "ok": false + }, + { + "task_id": "abc383_d", + "ok": false + }, + { + "task_id": "abc383_e", + "ok": false + }, + { + "task_id": "abc384_a", + "ok": true + }, + { + "task_id": "abc384_b", + "ok": true + }, + { + "task_id": "abc384_c", + "ok": false + }, + { + "task_id": "abc384_d", + "ok": false + }, + { + "task_id": "abc384_e", + "ok": false + }, + { + "task_id": "abc384_f", + "ok": false + }, + { + "task_id": "abc384_g", + "ok": false + }, + { + "task_id": "abc385_a", + "ok": false + }, + { + "task_id": "abc385_b", + "ok": true + }, + { + "task_id": "abc385_c", + "ok": false + }, + { + "task_id": "abc385_d", + "ok": false + }, + { + "task_id": "abc385_e", + "ok": false + }, + { + "task_id": "abc385_f", + "ok": false + }, + { + "task_id": "abc386_a", + "ok": false + }, + { + "task_id": "abc386_b", + "ok": false + }, + { + "task_id": "abc386_c", + "ok": false + }, + { + "task_id": "abc386_d", + "ok": false + }, + { + "task_id": "abc386_e", + "ok": false + }, + { + "task_id": "abc386_f", + "ok": false + }, + { + "task_id": "abc387_a", + "ok": true + }, + { + "task_id": "abc387_b", + "ok": false + }, + { + "task_id": "abc387_c", + "ok": false + }, + { + "task_id": "abc387_f", + "ok": false + }, + { + "task_id": "abc388_a", + "ok": true + }, + { + "task_id": "abc388_b", + "ok": false + }, + { + "task_id": "abc388_c", + "ok": false + }, + { + "task_id": "abc388_d", + "ok": false + }, + { + "task_id": "abc388_e", + "ok": false + }, + { + "task_id": "abc388_f", + "ok": false + }, + { + "task_id": "abc388_g", + "ok": false + }, + { + "task_id": "abc389_a", + "ok": true + }, + { + "task_id": "abc389_b", + "ok": false + }, + { + "task_id": "abc389_d", + "ok": false + }, + { + "task_id": "abc389_e", + "ok": false + }, + { + "task_id": "abc389_f", + "ok": false + }, + { + "task_id": "abc389_g", + "ok": false + }, + { + "task_id": "abc390_a", + "ok": false + }, + { + "task_id": "abc390_b", + "ok": false + }, + { + "task_id": "abc390_c", + "ok": false + }, + { + "task_id": "abc390_d", + "ok": false + }, + { + "task_id": "abc390_e", + "ok": false + }, + { + "task_id": "abc390_f", + "ok": false + }, + { + "task_id": "abc390_g", + "ok": false + }, + { + "task_id": "abc391_a", + "ok": true + }, + { + "task_id": "abc391_b", + "ok": false + }, + { + "task_id": "abc391_d", + "ok": false + }, + { + "task_id": "abc391_e", + "ok": false + }, + { + "task_id": "abc391_f", + "ok": false + }, + { + "task_id": "abc391_g", + "ok": false + }, + { + "task_id": "abc392_a", + "ok": false + }, + { + "task_id": "abc392_b", + "ok": true + }, + { + "task_id": "abc392_c", + "ok": false + }, + { + "task_id": "abc392_d", + "ok": false + }, + { + "task_id": "abc392_f", + "ok": false + }, + { + "task_id": "abc392_g", + "ok": false + }, + { + "task_id": "abc393_a", + "ok": false + }, + { + "task_id": "abc393_b", + "ok": false + }, + { + "task_id": "abc393_d", + "ok": false + }, + { + "task_id": "abc393_e", + "ok": false + }, + { + "task_id": "abc393_f", + "ok": false + }, + { + "task_id": "abc394_a", + "ok": true + }, + { + "task_id": "abc394_b", + "ok": true + }, + { + "task_id": "abc394_c", + "ok": true + }, + { + "task_id": "abc394_d", + "ok": false + }, + { + "task_id": "abc394_e", + "ok": false + }, + { + "task_id": "abc394_f", + "ok": false + }, + { + "task_id": "abc394_g", + "ok": false + }, + { + "task_id": "abc395_a", + "ok": true + }, + { + "task_id": "abc395_b", + "ok": true + }, + { + "task_id": "abc395_c", + "ok": false + }, + { + "task_id": "abc395_e", + "ok": false + }, + { + "task_id": "abc395_f", + "ok": false + }, + { + "task_id": "abc396_a", + "ok": true + }, + { + "task_id": "abc396_b", + "ok": true + }, + { + "task_id": "abc396_c", + "ok": false + }, + { + "task_id": "abc396_d", + "ok": false + }, + { + "task_id": "abc396_e", + "ok": false + }, + { + "task_id": "abc396_f", + "ok": false + }, + { + "task_id": "abc396_g", + "ok": false + }, + { + "task_id": "abc397_a", + "ok": true + }, + { + "task_id": "abc397_b", + "ok": false + }, + { + "task_id": "abc397_c", + "ok": false + }, + { + "task_id": "abc397_d", + "ok": false + }, + { + "task_id": "abc397_e", + "ok": false + }, + { + "task_id": "abc397_f", + "ok": false + }, + { + "task_id": "abc397_g", + "ok": false + }, + { + "task_id": "abc398_a", + "ok": false + }, + { + "task_id": "abc398_b", + "ok": false + }, + { + "task_id": "abc398_c", + "ok": false + }, + { + "task_id": "abc398_d", + "ok": false + }, + { + "task_id": "abc398_f", + "ok": false + }, + { + "task_id": "abc398_g", + "ok": false + }, + { + "task_id": "abc399_a", + "ok": true + }, + { + "task_id": "abc399_b", + "ok": false + }, + { + "task_id": "abc399_c", + "ok": false + }, + { + "task_id": "abc399_d", + "ok": false + }, + { + "task_id": "abc399_e", + "ok": false + }, + { + "task_id": "abc399_f", + "ok": false + }, + { + "task_id": "abc400_a", + "ok": false + }, + { + "task_id": "abc400_b", + "ok": false + }, + { + "task_id": "abc400_c", + "ok": false + }, + { + "task_id": "abc400_d", + "ok": false + }, + { + "task_id": "abc400_e", + "ok": false + }, + { + "task_id": "abc400_g", + "ok": false + }, + { + "task_id": "arc188_a", + "ok": false + }, + { + "task_id": "arc188_b", + "ok": false + }, + { + "task_id": "arc188_c", + "ok": false + }, + { + "task_id": "arc188_d", + "ok": false + }, + { + "task_id": "arc189_a", + "ok": false + }, + { + "task_id": "arc189_b", + "ok": false + }, + { + "task_id": "arc189_c", + "ok": false + }, + { + "task_id": "arc189_d", + "ok": false + }, + { + "task_id": "arc190_a", + "ok": false + }, + { + "task_id": "arc190_c", + "ok": false + }, + { + "task_id": "arc190_d", + "ok": false + }, + { + "task_id": "arc191_a", + "ok": false + }, + { + "task_id": "arc191_c", + "ok": false + }, + { + "task_id": "arc191_d", + "ok": false + }, + { + "task_id": "arc192_a", + "ok": false + }, + { + "task_id": "arc192_b", + "ok": false + }, + { + "task_id": "arc192_d", + "ok": false + }, + { + "task_id": "arc192_e", + "ok": false + }, + { + "task_id": "arc193_a", + "ok": false + }, + { + "task_id": "arc193_b", + "ok": false + }, + { + "task_id": "arc193_d", + "ok": false + }, + { + "task_id": "arc194_a", + "ok": false + }, + { + "task_id": "arc194_b", + "ok": false + }, + { + "task_id": "arc194_c", + "ok": false + }, + { + "task_id": "arc194_d", + "ok": false + }, + { + "task_id": "arc194_e", + "ok": false + }, + { + "task_id": "arc195_a", + "ok": false + }, + { + "task_id": "arc195_b", + "ok": false + }, + { + "task_id": "arc195_c", + "ok": false + }, + { + "task_id": "arc195_d", + "ok": false + }, + { + "task_id": "arc195_e", + "ok": false + }, + { + "task_id": "arc196_a", + "ok": false + }, + { + "task_id": "arc196_b", + "ok": false + }, + { + "task_id": "arc196_c", + "ok": false + }, + { + "task_id": "arc196_d", + "ok": false + } + ] + } + } +} \ No newline at end of file diff --git a/results-loop/eval_lcb_loop.json b/results-loop/eval_lcb_loop.json new file mode 100644 index 0000000..a3397e2 --- /dev/null +++ b/results-loop/eval_lcb_loop.json @@ -0,0 +1,1221 @@ +{ + "tag": "loop", + "since": "2024-07-01", + "n": 150, + "ks": { + "0": { + "acc": 0.18666666666666668, + "hard_n": 25, + "hard_acc": 0.0, + "per_item": [ + { + "task_id": "abc382_a", + "ok": true + }, + { + "task_id": "abc382_b", + "ok": true + }, + { + "task_id": "abc382_c", + "ok": true + }, + { + "task_id": "abc382_d", + "ok": false + }, + { + "task_id": "abc382_f", + "ok": false + }, + { + "task_id": "abc382_g", + "ok": false + }, + { + "task_id": "abc383_a", + "ok": true + }, + { + "task_id": "abc383_b", + "ok": true + }, + { + "task_id": "abc383_c", + "ok": false + }, + { + "task_id": "abc383_d", + "ok": false + }, + { + "task_id": "abc383_e", + "ok": false + }, + { + "task_id": "abc384_a", + "ok": true + }, + { + "task_id": "abc384_b", + "ok": true + }, + { + "task_id": "abc384_c", + "ok": false + }, + { + "task_id": "abc384_d", + "ok": false + }, + { + "task_id": "abc384_e", + "ok": false + }, + { + "task_id": "abc384_f", + "ok": false + }, + { + "task_id": "abc384_g", + "ok": true + }, + { + "task_id": "abc385_a", + "ok": false + }, + { + "task_id": "abc385_b", + "ok": false + }, + { + "task_id": "abc385_c", + "ok": false + }, + { + "task_id": "abc385_d", + "ok": false + }, + { + "task_id": "abc385_e", + "ok": false + }, + { + "task_id": "abc385_f", + "ok": false + }, + { + "task_id": "abc386_a", + "ok": false + }, + { + "task_id": "abc386_b", + "ok": false + }, + { + "task_id": "abc386_c", + "ok": false + }, + { + "task_id": "abc386_d", + "ok": false + }, + { + "task_id": "abc386_e", + "ok": false + }, + { + "task_id": "abc386_f", + "ok": false + }, + { + "task_id": "abc387_a", + "ok": true + }, + { + "task_id": "abc387_b", + "ok": true + }, + { + "task_id": "abc387_c", + "ok": false + }, + { + "task_id": "abc387_f", + "ok": false + }, + { + "task_id": "abc388_a", + "ok": true + }, + { + "task_id": "abc388_b", + "ok": true + }, + { + "task_id": "abc388_c", + "ok": false + }, + { + "task_id": "abc388_d", + "ok": false + }, + { + "task_id": "abc388_e", + "ok": false + }, + { + "task_id": "abc388_f", + "ok": false + }, + { + "task_id": "abc388_g", + "ok": false + }, + { + "task_id": "abc389_a", + "ok": true + }, + { + "task_id": "abc389_b", + "ok": false + }, + { + "task_id": "abc389_d", + "ok": false + }, + { + "task_id": "abc389_e", + "ok": false + }, + { + "task_id": "abc389_f", + "ok": false + }, + { + "task_id": "abc389_g", + "ok": false + }, + { + "task_id": "abc390_a", + "ok": true + }, + { + "task_id": "abc390_b", + "ok": false + }, + { + "task_id": "abc390_c", + "ok": false + }, + { + "task_id": "abc390_d", + "ok": false + }, + { + "task_id": "abc390_e", + "ok": false + }, + { + "task_id": "abc390_f", + "ok": false + }, + { + "task_id": "abc390_g", + "ok": false + }, + { + "task_id": "abc391_a", + "ok": true + }, + { + "task_id": "abc391_b", + "ok": true + }, + { + "task_id": "abc391_d", + "ok": false + }, + { + "task_id": "abc391_e", + "ok": false + }, + { + "task_id": "abc391_f", + "ok": false + }, + { + "task_id": "abc391_g", + "ok": false + }, + { + "task_id": "abc392_a", + "ok": true + }, + { + "task_id": "abc392_b", + "ok": true + }, + { + "task_id": "abc392_c", + "ok": false + }, + { + "task_id": "abc392_d", + "ok": false + }, + { + "task_id": "abc392_f", + "ok": false + }, + { + "task_id": "abc392_g", + "ok": false + }, + { + "task_id": "abc393_a", + "ok": false + }, + { + "task_id": "abc393_b", + "ok": true + }, + { + "task_id": "abc393_d", + "ok": false + }, + { + "task_id": "abc393_e", + "ok": false + }, + { + "task_id": "abc393_f", + "ok": false + }, + { + "task_id": "abc394_a", + "ok": true + }, + { + "task_id": "abc394_b", + "ok": true + }, + { + "task_id": "abc394_c", + "ok": false + }, + { + "task_id": "abc394_d", + "ok": false + }, + { + "task_id": "abc394_e", + "ok": false + }, + { + "task_id": "abc394_f", + "ok": false + }, + { + "task_id": "abc394_g", + "ok": false + }, + { + "task_id": "abc395_a", + "ok": true + }, + { + "task_id": "abc395_b", + "ok": false + }, + { + "task_id": "abc395_c", + "ok": false + }, + { + "task_id": "abc395_e", + "ok": false + }, + { + "task_id": "abc395_f", + "ok": false + }, + { + "task_id": "abc396_a", + "ok": true + }, + { + "task_id": "abc396_b", + "ok": true + }, + { + "task_id": "abc396_c", + "ok": false + }, + { + "task_id": "abc396_d", + "ok": false + }, + { + "task_id": "abc396_e", + "ok": false + }, + { + "task_id": "abc396_f", + "ok": false + }, + { + "task_id": "abc396_g", + "ok": false + }, + { + "task_id": "abc397_a", + "ok": true + }, + { + "task_id": "abc397_b", + "ok": false + }, + { + "task_id": "abc397_c", + "ok": false + }, + { + "task_id": "abc397_d", + "ok": false + }, + { + "task_id": "abc397_e", + "ok": false + }, + { + "task_id": "abc397_f", + "ok": false + }, + { + "task_id": "abc397_g", + "ok": false + }, + { + "task_id": "abc398_a", + "ok": false + }, + { + "task_id": "abc398_b", + "ok": true + }, + { + "task_id": "abc398_c", + "ok": false + }, + { + "task_id": "abc398_d", + "ok": false + }, + { + "task_id": "abc398_f", + "ok": false + }, + { + "task_id": "abc398_g", + "ok": false + }, + { + "task_id": "abc399_a", + "ok": true + }, + { + "task_id": "abc399_b", + "ok": false + }, + { + "task_id": "abc399_c", + "ok": false + }, + { + "task_id": "abc399_d", + "ok": false + }, + { + "task_id": "abc399_e", + "ok": false + }, + { + "task_id": "abc399_f", + "ok": false + }, + { + "task_id": "abc400_a", + "ok": true + }, + { + "task_id": "abc400_b", + "ok": false + }, + { + "task_id": "abc400_c", + "ok": false + }, + { + "task_id": "abc400_d", + "ok": false + }, + { + "task_id": "abc400_e", + "ok": false + }, + { + "task_id": "abc400_g", + "ok": false + }, + { + "task_id": "arc188_a", + "ok": false + }, + { + "task_id": "arc188_b", + "ok": false + }, + { + "task_id": "arc188_c", + "ok": false + }, + { + "task_id": "arc188_d", + "ok": false + }, + { + "task_id": "arc189_a", + "ok": false + }, + { + "task_id": "arc189_b", + "ok": false + }, + { + "task_id": "arc189_c", + "ok": false + }, + { + "task_id": "arc189_d", + "ok": false + }, + { + "task_id": "arc190_a", + "ok": false + }, + { + "task_id": "arc190_c", + "ok": false + }, + { + "task_id": "arc190_d", + "ok": false + }, + { + "task_id": "arc191_a", + "ok": false + }, + { + "task_id": "arc191_c", + "ok": false + }, + { + "task_id": "arc191_d", + "ok": false + }, + { + "task_id": "arc192_a", + "ok": false + }, + { + "task_id": "arc192_b", + "ok": false + }, + { + "task_id": "arc192_d", + "ok": false + }, + { + "task_id": "arc192_e", + "ok": false + }, + { + "task_id": "arc193_a", + "ok": false + }, + { + "task_id": "arc193_b", + "ok": false + }, + { + "task_id": "arc193_d", + "ok": false + }, + { + "task_id": "arc194_a", + "ok": false + }, + { + "task_id": "arc194_b", + "ok": false + }, + { + "task_id": "arc194_c", + "ok": false + }, + { + "task_id": "arc194_d", + "ok": false + }, + { + "task_id": "arc194_e", + "ok": false + }, + { + "task_id": "arc195_a", + "ok": false + }, + { + "task_id": "arc195_b", + "ok": false + }, + { + "task_id": "arc195_c", + "ok": false + }, + { + "task_id": "arc195_d", + "ok": false + }, + { + "task_id": "arc195_e", + "ok": false + }, + { + "task_id": "arc196_a", + "ok": false + }, + { + "task_id": "arc196_b", + "ok": false + }, + { + "task_id": "arc196_c", + "ok": false + }, + { + "task_id": "arc196_d", + "ok": false + } + ] + }, + "4": { + "acc": 0.15333333333333332, + "hard_n": 25, + "hard_acc": 0.08, + "per_item": [ + { + "task_id": "abc382_a", + "ok": true + }, + { + "task_id": "abc382_b", + "ok": false + }, + { + "task_id": "abc382_c", + "ok": false + }, + { + "task_id": "abc382_d", + "ok": false + }, + { + "task_id": "abc382_f", + "ok": false + }, + { + "task_id": "abc382_g", + "ok": false + }, + { + "task_id": "abc383_a", + "ok": false + }, + { + "task_id": "abc383_b", + "ok": false + }, + { + "task_id": "abc383_c", + "ok": false + }, + { + "task_id": "abc383_d", + "ok": false + }, + { + "task_id": "abc383_e", + "ok": false + }, + { + "task_id": "abc384_a", + "ok": true + }, + { + "task_id": "abc384_b", + "ok": true + }, + { + "task_id": "abc384_c", + "ok": false + }, + { + "task_id": "abc384_d", + "ok": false + }, + { + "task_id": "abc384_e", + "ok": false + }, + { + "task_id": "abc384_f", + "ok": false + }, + { + "task_id": "abc384_g", + "ok": false + }, + { + "task_id": "abc385_a", + "ok": false + }, + { + "task_id": "abc385_b", + "ok": false + }, + { + "task_id": "abc385_c", + "ok": true + }, + { + "task_id": "abc385_d", + "ok": false + }, + { + "task_id": "abc385_e", + "ok": false + }, + { + "task_id": "abc385_f", + "ok": false + }, + { + "task_id": "abc386_a", + "ok": false + }, + { + "task_id": "abc386_b", + "ok": false + }, + { + "task_id": "abc386_c", + "ok": false + }, + { + "task_id": "abc386_d", + "ok": false + }, + { + "task_id": "abc386_e", + "ok": false + }, + { + "task_id": "abc386_f", + "ok": false + }, + { + "task_id": "abc387_a", + "ok": true + }, + { + "task_id": "abc387_b", + "ok": true + }, + { + "task_id": "abc387_c", + "ok": false + }, + { + "task_id": "abc387_f", + "ok": false + }, + { + "task_id": "abc388_a", + "ok": true + }, + { + "task_id": "abc388_b", + "ok": false + }, + { + "task_id": "abc388_c", + "ok": false + }, + { + "task_id": "abc388_d", + "ok": false + }, + { + "task_id": "abc388_e", + "ok": false + }, + { + "task_id": "abc388_f", + "ok": false + }, + { + "task_id": "abc388_g", + "ok": false + }, + { + "task_id": "abc389_a", + "ok": true + }, + { + "task_id": "abc389_b", + "ok": true + }, + { + "task_id": "abc389_d", + "ok": false + }, + { + "task_id": "abc389_e", + "ok": false + }, + { + "task_id": "abc389_f", + "ok": false + }, + { + "task_id": "abc389_g", + "ok": false + }, + { + "task_id": "abc390_a", + "ok": true + }, + { + "task_id": "abc390_b", + "ok": false + }, + { + "task_id": "abc390_c", + "ok": false + }, + { + "task_id": "abc390_d", + "ok": false + }, + { + "task_id": "abc390_e", + "ok": false + }, + { + "task_id": "abc390_f", + "ok": false + }, + { + "task_id": "abc390_g", + "ok": false + }, + { + "task_id": "abc391_a", + "ok": true + }, + { + "task_id": "abc391_b", + "ok": false + }, + { + "task_id": "abc391_d", + "ok": false + }, + { + "task_id": "abc391_e", + "ok": false + }, + { + "task_id": "abc391_f", + "ok": false + }, + { + "task_id": "abc391_g", + "ok": false + }, + { + "task_id": "abc392_a", + "ok": true + }, + { + "task_id": "abc392_b", + "ok": true + }, + { + "task_id": "abc392_c", + "ok": false + }, + { + "task_id": "abc392_d", + "ok": false + }, + { + "task_id": "abc392_f", + "ok": false + }, + { + "task_id": "abc392_g", + "ok": false + }, + { + "task_id": "abc393_a", + "ok": true + }, + { + "task_id": "abc393_b", + "ok": false + }, + { + "task_id": "abc393_d", + "ok": false + }, + { + "task_id": "abc393_e", + "ok": false + }, + { + "task_id": "abc393_f", + "ok": false + }, + { + "task_id": "abc394_a", + "ok": true + }, + { + "task_id": "abc394_b", + "ok": true + }, + { + "task_id": "abc394_c", + "ok": false + }, + { + "task_id": "abc394_d", + "ok": false + }, + { + "task_id": "abc394_e", + "ok": false + }, + { + "task_id": "abc394_f", + "ok": false + }, + { + "task_id": "abc394_g", + "ok": false + }, + { + "task_id": "abc395_a", + "ok": true + }, + { + "task_id": "abc395_b", + "ok": false + }, + { + "task_id": "abc395_c", + "ok": false + }, + { + "task_id": "abc395_e", + "ok": false + }, + { + "task_id": "abc395_f", + "ok": false + }, + { + "task_id": "abc396_a", + "ok": true + }, + { + "task_id": "abc396_b", + "ok": true + }, + { + "task_id": "abc396_c", + "ok": false + }, + { + "task_id": "abc396_d", + "ok": false + }, + { + "task_id": "abc396_e", + "ok": false + }, + { + "task_id": "abc396_f", + "ok": false + }, + { + "task_id": "abc396_g", + "ok": false + }, + { + "task_id": "abc397_a", + "ok": true + }, + { + "task_id": "abc397_b", + "ok": false + }, + { + "task_id": "abc397_c", + "ok": false + }, + { + "task_id": "abc397_d", + "ok": false + }, + { + "task_id": "abc397_e", + "ok": false + }, + { + "task_id": "abc397_f", + "ok": false + }, + { + "task_id": "abc397_g", + "ok": false + }, + { + "task_id": "abc398_a", + "ok": false + }, + { + "task_id": "abc398_b", + "ok": true + }, + { + "task_id": "abc398_c", + "ok": false + }, + { + "task_id": "abc398_d", + "ok": false + }, + { + "task_id": "abc398_f", + "ok": false + }, + { + "task_id": "abc398_g", + "ok": false + }, + { + "task_id": "abc399_a", + "ok": true + }, + { + "task_id": "abc399_b", + "ok": false + }, + { + "task_id": "abc399_c", + "ok": false + }, + { + "task_id": "abc399_d", + "ok": false + }, + { + "task_id": "abc399_e", + "ok": false + }, + { + "task_id": "abc399_f", + "ok": false + }, + { + "task_id": "abc400_a", + "ok": true + }, + { + "task_id": "abc400_b", + "ok": false + }, + { + "task_id": "abc400_c", + "ok": false + }, + { + "task_id": "abc400_d", + "ok": false + }, + { + "task_id": "abc400_e", + "ok": false + }, + { + "task_id": "abc400_g", + "ok": false + }, + { + "task_id": "arc188_a", + "ok": false + }, + { + "task_id": "arc188_b", + "ok": false + }, + { + "task_id": "arc188_c", + "ok": false + }, + { + "task_id": "arc188_d", + "ok": false + }, + { + "task_id": "arc189_a", + "ok": false + }, + { + "task_id": "arc189_b", + "ok": false + }, + { + "task_id": "arc189_c", + "ok": false + }, + { + "task_id": "arc189_d", + "ok": false + }, + { + "task_id": "arc190_a", + "ok": false + }, + { + "task_id": "arc190_c", + "ok": false + }, + { + "task_id": "arc190_d", + "ok": false + }, + { + "task_id": "arc191_a", + "ok": false + }, + { + "task_id": "arc191_c", + "ok": false + }, + { + "task_id": "arc191_d", + "ok": false + }, + { + "task_id": "arc192_a", + "ok": false + }, + { + "task_id": "arc192_b", + "ok": false + }, + { + "task_id": "arc192_d", + "ok": false + }, + { + "task_id": "arc192_e", + "ok": false + }, + { + "task_id": "arc193_a", + "ok": false + }, + { + "task_id": "arc193_b", + "ok": false + }, + { + "task_id": "arc193_d", + "ok": false + }, + { + "task_id": "arc194_a", + "ok": false + }, + { + "task_id": "arc194_b", + "ok": false + }, + { + "task_id": "arc194_c", + "ok": false + }, + { + "task_id": "arc194_d", + "ok": false + }, + { + "task_id": "arc194_e", + "ok": false + }, + { + "task_id": "arc195_a", + "ok": false + }, + { + "task_id": "arc195_b", + "ok": false + }, + { + "task_id": "arc195_c", + "ok": false + }, + { + "task_id": "arc195_d", + "ok": false + }, + { + "task_id": "arc195_e", + "ok": false + }, + { + "task_id": "arc196_a", + "ok": false + }, + { + "task_id": "arc196_b", + "ok": false + }, + { + "task_id": "arc196_c", + "ok": false + }, + { + "task_id": "arc196_d", + "ok": false + } + ] + } + } +} \ No newline at end of file diff --git a/results-loop/eval_lcb_untrained.json b/results-loop/eval_lcb_untrained.json new file mode 100644 index 0000000..7ff2ad3 --- /dev/null +++ b/results-loop/eval_lcb_untrained.json @@ -0,0 +1,1221 @@ +{ + "tag": "untrained", + "since": "2024-07-01", + "n": 150, + "ks": { + "0": { + "acc": 0.18666666666666668, + "hard_n": 25, + "hard_acc": 0.0, + "per_item": [ + { + "task_id": "abc382_a", + "ok": true + }, + { + "task_id": "abc382_b", + "ok": true + }, + { + "task_id": "abc382_c", + "ok": true + }, + { + "task_id": "abc382_d", + "ok": false + }, + { + "task_id": "abc382_f", + "ok": false + }, + { + "task_id": "abc382_g", + "ok": false + }, + { + "task_id": "abc383_a", + "ok": true + }, + { + "task_id": "abc383_b", + "ok": true + }, + { + "task_id": "abc383_c", + "ok": false + }, + { + "task_id": "abc383_d", + "ok": false + }, + { + "task_id": "abc383_e", + "ok": false + }, + { + "task_id": "abc384_a", + "ok": true + }, + { + "task_id": "abc384_b", + "ok": true + }, + { + "task_id": "abc384_c", + "ok": false + }, + { + "task_id": "abc384_d", + "ok": false + }, + { + "task_id": "abc384_e", + "ok": false + }, + { + "task_id": "abc384_f", + "ok": false + }, + { + "task_id": "abc384_g", + "ok": true + }, + { + "task_id": "abc385_a", + "ok": false + }, + { + "task_id": "abc385_b", + "ok": false + }, + { + "task_id": "abc385_c", + "ok": false + }, + { + "task_id": "abc385_d", + "ok": false + }, + { + "task_id": "abc385_e", + "ok": false + }, + { + "task_id": "abc385_f", + "ok": false + }, + { + "task_id": "abc386_a", + "ok": false + }, + { + "task_id": "abc386_b", + "ok": false + }, + { + "task_id": "abc386_c", + "ok": false + }, + { + "task_id": "abc386_d", + "ok": false + }, + { + "task_id": "abc386_e", + "ok": false + }, + { + "task_id": "abc386_f", + "ok": false + }, + { + "task_id": "abc387_a", + "ok": true + }, + { + "task_id": "abc387_b", + "ok": true + }, + { + "task_id": "abc387_c", + "ok": false + }, + { + "task_id": "abc387_f", + "ok": false + }, + { + "task_id": "abc388_a", + "ok": true + }, + { + "task_id": "abc388_b", + "ok": true + }, + { + "task_id": "abc388_c", + "ok": false + }, + { + "task_id": "abc388_d", + "ok": false + }, + { + "task_id": "abc388_e", + "ok": false + }, + { + "task_id": "abc388_f", + "ok": false + }, + { + "task_id": "abc388_g", + "ok": false + }, + { + "task_id": "abc389_a", + "ok": true + }, + { + "task_id": "abc389_b", + "ok": false + }, + { + "task_id": "abc389_d", + "ok": false + }, + { + "task_id": "abc389_e", + "ok": false + }, + { + "task_id": "abc389_f", + "ok": false + }, + { + "task_id": "abc389_g", + "ok": false + }, + { + "task_id": "abc390_a", + "ok": true + }, + { + "task_id": "abc390_b", + "ok": false + }, + { + "task_id": "abc390_c", + "ok": false + }, + { + "task_id": "abc390_d", + "ok": false + }, + { + "task_id": "abc390_e", + "ok": false + }, + { + "task_id": "abc390_f", + "ok": false + }, + { + "task_id": "abc390_g", + "ok": false + }, + { + "task_id": "abc391_a", + "ok": true + }, + { + "task_id": "abc391_b", + "ok": true + }, + { + "task_id": "abc391_d", + "ok": false + }, + { + "task_id": "abc391_e", + "ok": false + }, + { + "task_id": "abc391_f", + "ok": false + }, + { + "task_id": "abc391_g", + "ok": false + }, + { + "task_id": "abc392_a", + "ok": true + }, + { + "task_id": "abc392_b", + "ok": true + }, + { + "task_id": "abc392_c", + "ok": false + }, + { + "task_id": "abc392_d", + "ok": false + }, + { + "task_id": "abc392_f", + "ok": false + }, + { + "task_id": "abc392_g", + "ok": false + }, + { + "task_id": "abc393_a", + "ok": false + }, + { + "task_id": "abc393_b", + "ok": true + }, + { + "task_id": "abc393_d", + "ok": false + }, + { + "task_id": "abc393_e", + "ok": false + }, + { + "task_id": "abc393_f", + "ok": false + }, + { + "task_id": "abc394_a", + "ok": true + }, + { + "task_id": "abc394_b", + "ok": true + }, + { + "task_id": "abc394_c", + "ok": false + }, + { + "task_id": "abc394_d", + "ok": false + }, + { + "task_id": "abc394_e", + "ok": false + }, + { + "task_id": "abc394_f", + "ok": false + }, + { + "task_id": "abc394_g", + "ok": false + }, + { + "task_id": "abc395_a", + "ok": true + }, + { + "task_id": "abc395_b", + "ok": false + }, + { + "task_id": "abc395_c", + "ok": false + }, + { + "task_id": "abc395_e", + "ok": false + }, + { + "task_id": "abc395_f", + "ok": false + }, + { + "task_id": "abc396_a", + "ok": true + }, + { + "task_id": "abc396_b", + "ok": true + }, + { + "task_id": "abc396_c", + "ok": false + }, + { + "task_id": "abc396_d", + "ok": false + }, + { + "task_id": "abc396_e", + "ok": false + }, + { + "task_id": "abc396_f", + "ok": false + }, + { + "task_id": "abc396_g", + "ok": false + }, + { + "task_id": "abc397_a", + "ok": true + }, + { + "task_id": "abc397_b", + "ok": false + }, + { + "task_id": "abc397_c", + "ok": false + }, + { + "task_id": "abc397_d", + "ok": false + }, + { + "task_id": "abc397_e", + "ok": false + }, + { + "task_id": "abc397_f", + "ok": false + }, + { + "task_id": "abc397_g", + "ok": false + }, + { + "task_id": "abc398_a", + "ok": false + }, + { + "task_id": "abc398_b", + "ok": true + }, + { + "task_id": "abc398_c", + "ok": false + }, + { + "task_id": "abc398_d", + "ok": false + }, + { + "task_id": "abc398_f", + "ok": false + }, + { + "task_id": "abc398_g", + "ok": false + }, + { + "task_id": "abc399_a", + "ok": true + }, + { + "task_id": "abc399_b", + "ok": false + }, + { + "task_id": "abc399_c", + "ok": false + }, + { + "task_id": "abc399_d", + "ok": false + }, + { + "task_id": "abc399_e", + "ok": false + }, + { + "task_id": "abc399_f", + "ok": false + }, + { + "task_id": "abc400_a", + "ok": true + }, + { + "task_id": "abc400_b", + "ok": false + }, + { + "task_id": "abc400_c", + "ok": false + }, + { + "task_id": "abc400_d", + "ok": false + }, + { + "task_id": "abc400_e", + "ok": false + }, + { + "task_id": "abc400_g", + "ok": false + }, + { + "task_id": "arc188_a", + "ok": false + }, + { + "task_id": "arc188_b", + "ok": false + }, + { + "task_id": "arc188_c", + "ok": false + }, + { + "task_id": "arc188_d", + "ok": false + }, + { + "task_id": "arc189_a", + "ok": false + }, + { + "task_id": "arc189_b", + "ok": false + }, + { + "task_id": "arc189_c", + "ok": false + }, + { + "task_id": "arc189_d", + "ok": false + }, + { + "task_id": "arc190_a", + "ok": false + }, + { + "task_id": "arc190_c", + "ok": false + }, + { + "task_id": "arc190_d", + "ok": false + }, + { + "task_id": "arc191_a", + "ok": false + }, + { + "task_id": "arc191_c", + "ok": false + }, + { + "task_id": "arc191_d", + "ok": false + }, + { + "task_id": "arc192_a", + "ok": false + }, + { + "task_id": "arc192_b", + "ok": false + }, + { + "task_id": "arc192_d", + "ok": false + }, + { + "task_id": "arc192_e", + "ok": false + }, + { + "task_id": "arc193_a", + "ok": false + }, + { + "task_id": "arc193_b", + "ok": false + }, + { + "task_id": "arc193_d", + "ok": false + }, + { + "task_id": "arc194_a", + "ok": false + }, + { + "task_id": "arc194_b", + "ok": false + }, + { + "task_id": "arc194_c", + "ok": false + }, + { + "task_id": "arc194_d", + "ok": false + }, + { + "task_id": "arc194_e", + "ok": false + }, + { + "task_id": "arc195_a", + "ok": false + }, + { + "task_id": "arc195_b", + "ok": false + }, + { + "task_id": "arc195_c", + "ok": false + }, + { + "task_id": "arc195_d", + "ok": false + }, + { + "task_id": "arc195_e", + "ok": false + }, + { + "task_id": "arc196_a", + "ok": false + }, + { + "task_id": "arc196_b", + "ok": false + }, + { + "task_id": "arc196_c", + "ok": false + }, + { + "task_id": "arc196_d", + "ok": false + } + ] + }, + "4": { + "acc": 0.24, + "hard_n": 25, + "hard_acc": 0.36, + "per_item": [ + { + "task_id": "abc382_a", + "ok": true + }, + { + "task_id": "abc382_b", + "ok": true + }, + { + "task_id": "abc382_c", + "ok": false + }, + { + "task_id": "abc382_d", + "ok": false + }, + { + "task_id": "abc382_f", + "ok": false + }, + { + "task_id": "abc382_g", + "ok": false + }, + { + "task_id": "abc383_a", + "ok": true + }, + { + "task_id": "abc383_b", + "ok": true + }, + { + "task_id": "abc383_c", + "ok": false + }, + { + "task_id": "abc383_d", + "ok": false + }, + { + "task_id": "abc383_e", + "ok": false + }, + { + "task_id": "abc384_a", + "ok": true + }, + { + "task_id": "abc384_b", + "ok": true + }, + { + "task_id": "abc384_c", + "ok": true + }, + { + "task_id": "abc384_d", + "ok": false + }, + { + "task_id": "abc384_e", + "ok": false + }, + { + "task_id": "abc384_f", + "ok": false + }, + { + "task_id": "abc384_g", + "ok": false + }, + { + "task_id": "abc385_a", + "ok": false + }, + { + "task_id": "abc385_b", + "ok": true + }, + { + "task_id": "abc385_c", + "ok": false + }, + { + "task_id": "abc385_d", + "ok": false + }, + { + "task_id": "abc385_e", + "ok": false + }, + { + "task_id": "abc385_f", + "ok": false + }, + { + "task_id": "abc386_a", + "ok": false + }, + { + "task_id": "abc386_b", + "ok": false + }, + { + "task_id": "abc386_c", + "ok": false + }, + { + "task_id": "abc386_d", + "ok": false + }, + { + "task_id": "abc386_e", + "ok": false + }, + { + "task_id": "abc386_f", + "ok": false + }, + { + "task_id": "abc387_a", + "ok": true + }, + { + "task_id": "abc387_b", + "ok": true + }, + { + "task_id": "abc387_c", + "ok": false + }, + { + "task_id": "abc387_f", + "ok": false + }, + { + "task_id": "abc388_a", + "ok": true + }, + { + "task_id": "abc388_b", + "ok": true + }, + { + "task_id": "abc388_c", + "ok": false + }, + { + "task_id": "abc388_d", + "ok": false + }, + { + "task_id": "abc388_e", + "ok": false + }, + { + "task_id": "abc388_f", + "ok": false + }, + { + "task_id": "abc388_g", + "ok": false + }, + { + "task_id": "abc389_a", + "ok": true + }, + { + "task_id": "abc389_b", + "ok": true + }, + { + "task_id": "abc389_d", + "ok": false + }, + { + "task_id": "abc389_e", + "ok": false + }, + { + "task_id": "abc389_f", + "ok": false + }, + { + "task_id": "abc389_g", + "ok": false + }, + { + "task_id": "abc390_a", + "ok": true + }, + { + "task_id": "abc390_b", + "ok": false + }, + { + "task_id": "abc390_c", + "ok": true + }, + { + "task_id": "abc390_d", + "ok": false + }, + { + "task_id": "abc390_e", + "ok": false + }, + { + "task_id": "abc390_f", + "ok": false + }, + { + "task_id": "abc390_g", + "ok": false + }, + { + "task_id": "abc391_a", + "ok": true + }, + { + "task_id": "abc391_b", + "ok": true + }, + { + "task_id": "abc391_d", + "ok": false + }, + { + "task_id": "abc391_e", + "ok": false + }, + { + "task_id": "abc391_f", + "ok": false + }, + { + "task_id": "abc391_g", + "ok": false + }, + { + "task_id": "abc392_a", + "ok": true + }, + { + "task_id": "abc392_b", + "ok": true + }, + { + "task_id": "abc392_c", + "ok": false + }, + { + "task_id": "abc392_d", + "ok": false + }, + { + "task_id": "abc392_f", + "ok": false + }, + { + "task_id": "abc392_g", + "ok": false + }, + { + "task_id": "abc393_a", + "ok": true + }, + { + "task_id": "abc393_b", + "ok": true + }, + { + "task_id": "abc393_d", + "ok": false + }, + { + "task_id": "abc393_e", + "ok": false + }, + { + "task_id": "abc393_f", + "ok": false + }, + { + "task_id": "abc394_a", + "ok": true + }, + { + "task_id": "abc394_b", + "ok": true + }, + { + "task_id": "abc394_c", + "ok": true + }, + { + "task_id": "abc394_d", + "ok": false + }, + { + "task_id": "abc394_e", + "ok": false + }, + { + "task_id": "abc394_f", + "ok": false + }, + { + "task_id": "abc394_g", + "ok": false + }, + { + "task_id": "abc395_a", + "ok": true + }, + { + "task_id": "abc395_b", + "ok": true + }, + { + "task_id": "abc395_c", + "ok": true + }, + { + "task_id": "abc395_e", + "ok": false + }, + { + "task_id": "abc395_f", + "ok": false + }, + { + "task_id": "abc396_a", + "ok": true + }, + { + "task_id": "abc396_b", + "ok": true + }, + { + "task_id": "abc396_c", + "ok": false + }, + { + "task_id": "abc396_d", + "ok": false + }, + { + "task_id": "abc396_e", + "ok": false + }, + { + "task_id": "abc396_f", + "ok": false + }, + { + "task_id": "abc396_g", + "ok": false + }, + { + "task_id": "abc397_a", + "ok": true + }, + { + "task_id": "abc397_b", + "ok": false + }, + { + "task_id": "abc397_c", + "ok": false + }, + { + "task_id": "abc397_d", + "ok": false + }, + { + "task_id": "abc397_e", + "ok": false + }, + { + "task_id": "abc397_f", + "ok": false + }, + { + "task_id": "abc397_g", + "ok": false + }, + { + "task_id": "abc398_a", + "ok": false + }, + { + "task_id": "abc398_b", + "ok": true + }, + { + "task_id": "abc398_c", + "ok": true + }, + { + "task_id": "abc398_d", + "ok": false + }, + { + "task_id": "abc398_f", + "ok": false + }, + { + "task_id": "abc398_g", + "ok": false + }, + { + "task_id": "abc399_a", + "ok": true + }, + { + "task_id": "abc399_b", + "ok": true + }, + { + "task_id": "abc399_c", + "ok": false + }, + { + "task_id": "abc399_d", + "ok": false + }, + { + "task_id": "abc399_e", + "ok": false + }, + { + "task_id": "abc399_f", + "ok": false + }, + { + "task_id": "abc400_a", + "ok": true + }, + { + "task_id": "abc400_b", + "ok": false + }, + { + "task_id": "abc400_c", + "ok": false + }, + { + "task_id": "abc400_d", + "ok": false + }, + { + "task_id": "abc400_e", + "ok": false + }, + { + "task_id": "abc400_g", + "ok": false + }, + { + "task_id": "arc188_a", + "ok": false + }, + { + "task_id": "arc188_b", + "ok": false + }, + { + "task_id": "arc188_c", + "ok": false + }, + { + "task_id": "arc188_d", + "ok": false + }, + { + "task_id": "arc189_a", + "ok": false + }, + { + "task_id": "arc189_b", + "ok": false + }, + { + "task_id": "arc189_c", + "ok": false + }, + { + "task_id": "arc189_d", + "ok": false + }, + { + "task_id": "arc190_a", + "ok": false + }, + { + "task_id": "arc190_c", + "ok": false + }, + { + "task_id": "arc190_d", + "ok": false + }, + { + "task_id": "arc191_a", + "ok": false + }, + { + "task_id": "arc191_c", + "ok": false + }, + { + "task_id": "arc191_d", + "ok": false + }, + { + "task_id": "arc192_a", + "ok": false + }, + { + "task_id": "arc192_b", + "ok": false + }, + { + "task_id": "arc192_d", + "ok": false + }, + { + "task_id": "arc192_e", + "ok": false + }, + { + "task_id": "arc193_a", + "ok": false + }, + { + "task_id": "arc193_b", + "ok": false + }, + { + "task_id": "arc193_d", + "ok": false + }, + { + "task_id": "arc194_a", + "ok": false + }, + { + "task_id": "arc194_b", + "ok": false + }, + { + "task_id": "arc194_c", + "ok": false + }, + { + "task_id": "arc194_d", + "ok": false + }, + { + "task_id": "arc194_e", + "ok": false + }, + { + "task_id": "arc195_a", + "ok": false + }, + { + "task_id": "arc195_b", + "ok": false + }, + { + "task_id": "arc195_c", + "ok": false + }, + { + "task_id": "arc195_d", + "ok": false + }, + { + "task_id": "arc195_e", + "ok": false + }, + { + "task_id": "arc196_a", + "ok": false + }, + { + "task_id": "arc196_b", + "ok": false + }, + { + "task_id": "arc196_c", + "ok": false + }, + { + "task_id": "arc196_d", + "ok": false + } + ] + } + } +} \ No newline at end of file diff --git a/results-loop/eval_panel_base_k0.json b/results-loop/eval_panel_base_k0.json new file mode 100644 index 0000000..ea49778 --- /dev/null +++ b/results-loop/eval_panel_base_k0.json @@ -0,0 +1,8 @@ +{ + "tag": "base_k0", + "k": 0, + "arc": 0.36, + "winogrande": 0.55875, + "hellaswag": 0.5225, + "mmlu": 0.30125 +} \ No newline at end of file diff --git a/results-loop/eval_panel_distill_ff.json b/results-loop/eval_panel_distill_ff.json new file mode 100644 index 0000000..6728ff3 --- /dev/null +++ b/results-loop/eval_panel_distill_ff.json @@ -0,0 +1,8 @@ +{ + "tag": "distill_ff", + "k": 1, + "arc": 0.4175, + "winogrande": 0.56625, + "hellaswag": 0.57, + "mmlu": 0.3175 +} \ No newline at end of file diff --git a/results-loop/eval_panel_loop_k2.json b/results-loop/eval_panel_loop_k2.json new file mode 100644 index 0000000..dec0f2e --- /dev/null +++ b/results-loop/eval_panel_loop_k2.json @@ -0,0 +1,8 @@ +{ + "tag": "loop_k2", + "k": 2, + "arc": 0.36125, + "winogrande": 0.5525, + "hellaswag": 0.49625, + "mmlu": 0.3125 +} \ No newline at end of file diff --git a/results-loop/fig_transfer.png b/results-loop/fig_transfer.png index 19e427d59ed20d0e66e483f78967931a6cbd74b1..c470b14c8d98d4e33ee92affe2aeaaddb61084ed 100644 GIT binary patch literal 43841 zcmeFZXIRu%+buqcF-Ap<6;MHqpn?hligb(x1q7rwfzW%CGIUH-h=Pg(f=E;89i(?7 z3L;JEFbs$Y2t$#sAn;$?=Xu`uf6k}(%Q>Hp*A)wancwVv@B3c&TI=4oE-PKwv|-l< z3Wc&s?&3LB3T15u3> z%&t0HIoJ#H3kvcdKf2G-#l_KCTtL9?zrKUt;i`qegPH^fyvlmVi#pB}%8tLue?PqJ zIQ^JHNzITu_q)1#)EI;6zFlK!bxL{H{`K6y>VAs8n6=?a_`ZGnc(i{_V0*Q4ZNr9X zpY1z#>`;%t_1Cq18n^%6KEk;+o*IAit=vBD`1;AszNMS%>}d_Q`*yomW(F!a57kF_ zjdqn&hk^o(*!}VT-+!Gh@i2G$&p#-p8r)q!{QF$bki*Mh>T^M9C(`Y;j=GI4RYiq?_J_VPRmunY>|xC z-MD&j3&x19lZ>P-Gcc$|&@^!T2g++cNtUYPLK44Fo>vXOxVf>!Wkj`Pc;lu`4*~*e zr2hWYEAIUz+jD*_Wz%kv*UfXjGYzLa2Onq&i3vI1-6Qt&nOPm*i4*GUSogec)y_01 zPwgpoPCNbQkK)vcD~Z)KTKcaCdSZ{%sF+9RPPk+g?d2+%J6}D({_NE}g<`s+2kY>9 zyP@~gP3h&?7MCXK%#fm4Q_M2gIhRRx|0na~bep~kpXs?djr+fy|2RILfmgS1vt-2z zjC}5|nx2|c8EH-sme^zZ>pb)Nk=*%XHR?mJ>&I8JC!MuAWYX*^({~~^Bgn4pQ~THy zYxv~J6Su*ddwV-F^a|CJFYh0$k7%(kNwR47ihg~UQzFlyPkINR`o&VWan;^34`GRT zb^Rj8-v{f$wfbM&bX}Yssb@W;XkcKlVbdm!3h%E%659*I+UAdpCAg$pv{wv1e6uH_ zn6Y=>XJT~GB7Z!*?ZaBiK4B6lrngtW4a+Px`K_2}t-L;P@6-E#t}lKF6qp^$Z8AS& z$}!oQV?vA1&OTnKoNaMl%z5bJmaCH6McYZ_Es&53=82N@P|eB7;WH>J`u6P`w;Y>* zL2H6SxO)7>Eo*=Jsk8Ooy?Y&9xfVNq{q^y;uQN$$jG5sk=1@Z<-<6{5vU0UazHQsK zxz3O2czAlIUVd^$gxXzGT|N4-HbmU_TM_q2dxl>7Pde833=Hv!1qB89E>n?l*O)obkwdOCedu8U8se))3n^v7D0Q_ItXb>e<2 zWq?7FXO+h|X{Fl=+?%wPEM4}yc=XRpdJ7fsw^EoLi9BO`2{9%duyO zn~Zhl$kx($Hp%&Oo8V7>Blhb12QCEte*I^T1o;O-Sl;v&Y^y`rT)STNo!zI#!l)Ld zA%~PjvMGIIoMpZxl{d&;Lm*+~qBA(4%Mq6uN$4HgtS0Fl6l9V_Hl45;@aN;#H zbj_mO7AkS%>3MA1wwaVI^m@r6$TM8a%F2T9+Pn1%Y@LcSqPh;MsHjx0EHB+rOHvZc zUtY(;9&B9y!w)~Sdj@RYE5glzGgbZIh*pra&m#NE%1V0)w(yi@y7v7KSKY=sjU>N* z3=yFVXk}ihw#h0tKX>7R#E!IA_yHC<**||=Yx`z0XbRyVJKPw($JC2A zE_`^eg^82+XGYB(JpQ{3y;BE7L_}VR#65i)WD)Dr{yOYfWPU?;zRm7P5xcA6vQ@`} zZf<1T6)o;6baklS%C4rRr6pM>H&M`U#k+LTyr4XH>Ey`;V=vwM2;rJ!gYpu^D`Zt= z?{V32A3CJR%+N1w4>k=wqQ(D8^xj?>dv|wt_TPX1y}BXtlwD$k*o5!bLAE%gU6CBb zEC-{Mddh9GH+sme)N?Mie{j(3r*-Rg%gM=!v`ByZeAXsX!fh-&qQI)<;91(cA}14@ zh`51IzR|hVy4xKasvLd&_zf$1f=%=68Qn$_rPN6*rAC@&KJO{7JB}hE!t`abDjk{n z8T4`-*LKg5;+6OXaoW2jex6J&lSYGoN7+FsW@N8#{qZOnw#gnYs)YOJ$8*l~(#ZUG<7WJc#Remj4g-PhN*{cEo$^H`(@&b{$=KK0k)Qic`YvbVPKut`~VpEu?U(?8C27%HDsaVb=mS#CY7mKe0Z_VfIFLG^B&k=A6DD&3~(ghK5cAS@Jl$VQeqQD2x%JGIH#jgkZsWR1 z#wS>+e%Y|$*+k`EYd7w?xnYM8s%nPdrAH@fjaWvy<{KwET6+S=t=RpirfT-;B`qy3 z?&47hJMrU>Kc<`Mm$>Al7?!wL;b#qvfO$*vA^X;wf&?nK7 zy~F7Tbf$Fg$1IOn)1=Bfez7o7m~<1+%}Z8EQRQJ?tqVJ5bHeiNb6o+cj@a2N7lb_f z5~?4V9F^+WtCba}Gae<^w3t}9SD~r+p(@R%%72q=jHCyPfPjF>w=Yvl8QV~I2xhsa zCwHWlzt(-OP3w+LM`l-pgnf_b05hvn|P_8iP&0P1gzV7A(!v9d@I7 z;PA>C8$X@u^_tYF>@leHWr#PS)NGQI#WDt6U$x4 zly&HO`PL8eB@Z5~W*2pQyq{4qBPA^@&7gCa%emovzkYp@DWqonwfP96;>dJipKosO z6m|NeoQ?5gi>4TnnP_RB?$`GZ1njTDtG>+Hk(Q+0&EpbNO0m-8*$s~$?=wNLw$Dksj)uNV z*AYHTQ^9^l?sZXBc(yRs)uLTs8?l9r&Ei<#ty_W>i=#)X>gwuVN1n=Rs5n(R9^I9d zQp`qoMq+D!Ik4NI%t&h0o>5{J#3j}Jd3qXWQD*g9t}9m^qt9n)-jzLxXT9Gr9{&{Zh&Y7_AsCcL2b=mKieDr{=4v6 zPt|gE$u5Ra5Z1Yo6~?bNoTW_xM|Fnk(e7}HyEt}E+}d`CboMMw5b!j>&Ub0N&~4^Qk>fwVu&|uq=BQj<@p>fKI}Vgt z{p4&wdW{__$=KFv^FkllgI&VbUh|WEeZi*A7UR*_A&ZMbNH!M7Bs`|$`$tAb97P|x zGTX(<2HuT1S;)yVzE)l1IHHw_*G2g#M1m}+f8_4antZ>1a?(NZjLz22KzS#&Zrz$b zW2#P*lB!X(e%77;o79r<0ovF|v26UIf?4~8(xsMAQRihR^R5MBoZm4)^PxrO%8;=L zpT>teh8B0@X7@}|4|P?N;6t@LOh8-9?*P|{tL4#~DA$g@FCLCjaMD#(rJ>AaY}mnf z`1tY5sCuejrw0LHJvOl)KTS&)s43B_xyw2JhQK?d54l1^iut?8zpnvUwEpn&)*>>r zE6{lhb7nZ!@6Bwk+r60se zZ~t@6x)W|=+AV^K2eAhmHg7IGlBJq(o0~pqI<962kZJr}?d}(Cqgv z-MV#)7XS2oV5(fm!4kLeM?+1q4r+clb+q^iMkzfsQfgw?No#I7&w)FqW}9SIMT6dP zpmv7D#>Td18D@R?;vV@BFl7f9S7v9YHgmo++vxg-fq^!h*%O8pQ3eqnJ*M;dzs1cb z2{(J)DbmXE@bY4&B*n-(m2@$ct%U}}1rk=-g`Jm+V|)6eoHRY}9T2YZx_b3lulI~% z`PZRHKD~l;5xegAM<;E}KY3Cp?{5b!x3W^6OKCnmcWZkRo<%_|@zQ;6;O-Q3UmJe; zB{(T530T_|tHCI7t?MrXFmltL7#=<(Ct{e5RZkZ*ss5Bv*sB`A%K33%Anx{#BdN$v zWF@NvUhyI}gSf56yYunZwyCZH`XwK;CG+jBW|8&?XVq=+Zb9U1<(v3Mpn;RwKblsWtz8- zk4Bb(REBoW8RSUykb{bONF{66txNNq8!f?m`m=JT?l_{E>@-lVn`3-4o?iPbC@i8z z(cJKekhP!qf%1;Oy>{rKDh%S6P;X)|8NwBZuSy*#tD{1$M$m z@3I+ss{lj1+w1r336Q=uzPh-wG&#qh_VG$eUa4H3*Snq5TiY@-f7Yznc~}$qTrDHA zna$fGY=HkLQ}3B+?TLZ;sj1?<`y;mQ-W@q~!s5eTv8(YY+v;hW#}s2Eb?`y+B6=%) zypCySKhOVz@oj zU5Csvnp4B2_QxNO(JVXxx}o>nU(7CH>$n}%6Kl21e!th|oNTpQQfmRDyky1n0kf=& zd9UzPi1mE_;qbdH$WU+5A)8Wj>UKw=+#L z4yTq4ST+UYiE$$2kk)D3`o&kz<1ft!{1ZkmeoC4+nA`JCyVT8Y8LP~}6~}RRBC75% z8tnGkkVA7{hhvZY_S=V;W5WieR|jtcAY=d>cPi%Yv1#lvvkytkbG z>f-9K!Gu<>eCXjP=wz&cI;}gipD<>lC8+Wx%Lhmko8Qyaqzt^9SAnXd7%i>=fbvZG z)zyW|f;46g9&S)Q^NJX=Z?}l;<(Z*|IiM>%z!`j9K6jJC3-7@<4^q+E&_Y5o%s@yW z*6)WM)p@$@kkY5g%2g5#8C4nyJh}6Bn&q0JxiZJ7{RS1@T}HdHUGlCR=2=@BUxumB z8TE5D5oQ)x+z95qG`e(Jv~c4~ro2p8j7^%^Luo%hKOay&ZgRd zkLTv$=8l(WC)C9AKi7!+EatV$Xmg#f^qjLU2@l^{*ST;&Hs~$lT{S(tS;XzoUf(ZI z^t%e|=;<ZtP1XTJ4yI{bLW(BtABO)U+a97ZV4RcMh!W5=n(VzX6az=lMcN- z{9P?^7dFc^hw`eCW}k!~nmOyfigL!)dM+w2X8`i1+LJRkx9{DXg#woJ%zsra*IaS1 zFN%k$6V6jMCWwnm}vN(Wr?lO;D@M>yus+N|?YIcLQAs5d2v%EuRn06?} z{fvI8n+CHzLj&CuQ*;{Hur*yLHxK*3tUC}rhknLp_R~9SMoIoo0fVtt$v^-6^G&Qw zWn-*N)@JE1r|swk4^aM)-Rverh2yA`y}M4?N1!fQ7Y9wF&eQADN4HU%*)7QU}`g!5*o_9#^PsZtmNhRsC1_zqA>B-trg**0I zgpI$c5N@%EFdJGN7%gbB9u5+odL(xDN_oLuc0qEOc8S8|P)5~393q2vcAZ3Tfvr_=>eVvvcol5A@SPYUVEUpEAVgTETdcrc%3rajMUIAiKe!+*2&592LJBD;{-QkdN=pVUV3wZuq>Wgvpxlb~K|W zNGRQY()`J<)`p+KtXUH;p+wY~lTN zFx;Oz4sk6q^RQet{*;JTVX4u>(iPEM+p6Q0lL$|Ti$XCTLlN5tDk6{6JVYiv(xGB8 zL``z8j&N_-b74fRSB=hf;KE%r1Hh%#NN3IY1_0?N$Pq7kpqd~rSK_nK8jV_%2%6&S z(o&M>Bo-yEhRf>`fDT)R`Ff59^s5Uvu*wK70|;!Uy{Mc8=fdy9=HxUx!b6Oq<#< zN;Jr(19qwBS*o<9X(=m9&4n@om7}{3uHCTxGSUb_{gH8W=m%cV3_FkM#)H+w0oDWH z%vNdw8;8x$SYq18#KgGK8=j&iB}oWyIt%b~6oGd8-aV~<{#vV%s?I9~UT~x&u%oU{ zfw1!QLmBKX-2!2D;SyB)fELn=owVy#7dEedi~74``}TO`j*$5aSzpbkhBR-gW{w0-zHodMYm-|3>=oK@v^3yJiqqC^uj{Nr+0A#E7%|vq;Snp zC_M^X;9?>;f1*%MYdU^>z2yOL6BayZ%|;5xrw^|JNmdtk9gWw?H8*n;h1@XmB2f9e z;1w}SkOX#J8ilu>buERmkHOAQG(Spk3YXtEr}4f@ncOP;^Y~4P|4k#i782zVv5$P~ z`~LsD9{I*w|NA$RZ=q2B?}=&uN0qw&_u~KB&i}We{}Wvo(ZXd-I6x$f*@Zqo3%A3l z3A|Zl)3?z%I5mx~p>Wxrv~GRt+$0s=Wz?f4v+^Yl-ww`sc;;?>69wf<-{dR-^ z03gmgVwnBMp8#U%#;k#{J6oJy6B z{eR(gh0?EIy%1&>++4XdbzV|LL=(afN^HB7;ZDjvXVa#dYR9M9(!wlV%-iZv1Wcs zUWuOazi}hkvHt}aqelk5U(JwfUl{3YZQX(LAUBs=1E$5c#E$aZ4tdipwNe7K4l3c?g8pYuo(8@N8kiVxN+krwV6})J)McMD0soC z8fl5Boi<>KIQC>P(F(f&jfB&?@~lF9zx1!01HqT*4Z@btAf!@2^5W&=lLt=a7f`05 z%S{9(9YKq6SzTS-KkrsRAk`coJ^jOn59IBDqD7-(LPJB1I73UH*{#Ehiw^j*#REk$JLfIt#kWq5{3oZxsrZ+Xxm*Dc``wB@fT{5xV}= zKnHV>-!@?m4ZRPV@GS1zhkI3&d~E0o&O!kZlvG`N}+z4`@57Cdysw8^A&zpDOltyAn6Je-hk@TQV8d7%QmVC zefZG5?}L+noE--TM+%a#1LLOd6WiSy0scA}*p zF}M+t3ch~F;mgna^U$>Ok=Ig=X$7KEM^8_h{k=Z3D*iP@Jqn1GfOy-6Z(}j!qCC$5 zJz1MfiPOdpnpduM(Q;nDfQ0epFiy-|8ivN3`TRi38mX|Y@Lk9ya z>q@+_FcEw7t0mx+0F%-Gx#l_`n*t5bmTk)uj+nP z^(VJP@T)SsERLd+dy)ofEx0bDWiRN!bIFx5UV&131AB?L5H}_O)nsrZ6j`Y`$+w?1a^=aF8?o31jdjyh2 zVipm2)C2{zvrpfXSt)8!_L!Ss=)olLNW>0%3MBO4l!HNlNmL>=dNQnCJW=GJ7nG^= zvtLfwv|j?XcRzl+j6~6ky1FB;`if$6m%S+=6FJ7!SUnW%&lL7Ep2cXvyC^DYUu^hYPGo^IulPewctmTP6H zZ?$`q+Cj$3mLj|;pGImZ=^K$i=spJ755-+mfqm1+517@3b>~^B-DXZ)>e&f)wB0uq zN+{V2(tv*t%8<#%0y_?wWkkd%qqOi%*s(OkruFOK{j9OO=z!}WAQSMFRF{!vaD5rr zrJ>{I7IYU=AUSl!%F#sAo-o>BHT#63cH{PVf|8J0ZPRS?UHoQ zW;Fav&7^`~^<0?9K(ml<`R;{yInFY%_}pgx9Dk$V_}bCx>0}o)z>K`ofbg84RT1mf z!$_Tty1tNdI7J;#gTB%%_nb@TY(f^=v1JSCSqWF#z6gX~N_vV&Gt%EHgdKe+Q2Fg? zQjF&t`v@TAHUQC#L2?8UPx;t`2YecQ`plkGjU%A#)yP59-vdsw1&#Ea-|B*2BJs+g zGp3CN@RqtyJ*}$g+_rVAt-p65dImxv!kc3gjWl7|@aQDb9Uu+FgH|{!Ep0%v*< zNzJvYiTI3QG1Bm&BXoz#{CU44n(1#5?rmUJBVUd)V5qQNSs99SnC-U7>I&tQ_SQz8 zY(XU0ZOtWg7aAkG585~uRcxi=yQ%Ut8tsHbuVl20U+=_)nb|x}lon*e6eLq|$M@Ir zz2S<&F47vd`BtthrlIxZh@?gU09-<6(~hhahd9-%%QCE-P>0PLG`B?pwN-z@XybiQJ<_+(Pbh6o-BRE}u`1P?sqp zZWBEzc#{+`5Zy&iaVyIUh8c#H27nRyzE*Vfs5rhRGp})6Om9=i7*^=1r)Ou?L87$m z(5m!ZCaRCN;k8Aud!K+}M!^p*YBjIVOdp5_BuOXX)WyLyoT8SbS{Aa5SmdaX zq+D|bggQ7fPc4(cei}f&R6BR>M!w;lyLU-GAB3w74j=1<35K}mY%+0}fR(=lZmJb6 zGdE)*DS>{$I(i{9;SePYnRAvijhhKj5GvG1U@$F z`*whkCz<1*a*S<@z2}F5qM7wMKd#;Iy06a=;(`i3;9SG0zM|H7(%@bxzG?;9gFMJk zV|1FN$MmOwy}p-$MB?d_dT@wE!eBLrowf#mB1aI^U^-}kP6p-ErO1>n^tA{)3pGtB zi05Tx`F{KDMTs9O?Md*pa>P(KS9>Vd5|Kq5XYHB#X^4{1tjd*VUa877B)mja@MHQV z>LhD`xyfkRu=8j#D2#A}yt@J0sDQ0U4<3|-yI{eE16$YTw=%Pr*#{0l9ppLTM{r#0 zJZrk~(8(Oa=9X9xk(om4wl@=jK9D!-{-|@ z^7QwOrPHOZ%tUZdqu5s!>^?VR<)jz#ZmMBrMFV5X9GOZ|^61&Hmo+tder4bxGT@N3 zN|mbn^#DvkTH4wVxl>?Qf&39O7ifXc4R@xE&qoQVo%~ml(Q6+DiXC~Bg#$pmc_csK ziY~4ZWPQT;=e8B5mywcS(k?Gl_g)JjCt0 zciST!DpF87n#CPZ$=XU>ZA!UIh3yi##l&<79R=Hx?VIxQ@*P{ZJ_V9dpfQGn&Ij)7 z3TiNbTj%D@o1_o|T^B9|vBCNX|4sZC4<1obZG32DL8a$gXk(`j62v*ztRT?!0Pe>9 zRTJiwORPc)XG)M!-R|88=hKjf)hwTPW_Gp}?}4m!0^Ein%q?0dj&ob`0O;sE8I#^Y zXaX#nGJ#OVzYU5f*!?kbI)Q#`26&eVW8JfZU10I4#4;M|J$PHF}7u8SY7%WZI zcR87}?!(kHfC6%C!wo!%_^plw-vM+4VZP{_Q0Bvj>=_2-x_iAp|6VdYH$AP25T1L* zW|#~mjZkU6D_nz2!RM)^ks4{-N}j_iIY2nehoIKO!=p)~#JTf^s6_3RBTj zM5b7jOu39Hg(9R0hvqDxq7Am9eKJ&Lc{T)SBN0VZo2g^ZvitZS*XDR0?+lC2G7EC8 z%XYk5J*;$Vu;vpsi>?T5|2l~RL`DjtfOvxA=`x*twbR?XqvhsjU)sUckOdou?kZfd z&%g*K!M#F6RN@E#R#wbk1}@YM>E9Ve7fy({X;EQl*8FDu(~muIqG1lBT5l)RH5v*BtzXZ9xa z9r#gh85x;U!S;ri8!7JYQDFTlDk{Ev0U9EcP`y$K#Dl4!VRISE4C@kIqUJz8J0uSS z&k>$uI?v9L&It|($khCI_R1K*M|Yf)31o&teyqjhuX$ixG$8j0O{{L{Zhb$-wecc; zMgxsG>6jire!RUVNV$&aNWSNsT8H%vyf;uBWtyZGqHy4oV|^Fp;3MhiE%(YdsbM93 zcKhUu=GDanl>Efbu)%|A$e0Q=EWHIM?clMn$(MJ1uN8W0X1C^i>rm-)UHCR~i~eEa z&`W~opPX5X4nm$h5ZPAZrBtW$`Z$EIgg-1y?~9JTuR3rkl6ML#1vc2_>$F)?7!RZ3 z_Z(P#cq;EghwN!FG{Mhe})#-EN9 zuZyat4#)UMd{($}bk_XLQsiI~MP;@tB5c}GEa$3sJ6M91Agwn9A%>6v<(d~UA(I0a z9Q^AhE=g^4ndJ+;En0>?X2PvHc*k-7%OR!<#>WED*{328(d_+h1^xZ^Wf-{8kMqC| z1eqxD>)^CK5uhT3%+lBYHOr@&{!q8THU+7|5+s6EZ$@mdYei&)YpA`9Zp@Zjli*pZ zh7#9k#d3o`H3Vh=w~%oHCQ{POd^cJ|_AAG)N<~Y-N7R;~*I}d=CeeusN*u;k$ibeF z=c46fcA}CukB2+8%fu!Qd?2}8P&BYkh zC5@TR^3A1pIT~o`h2IpKzLdg%_!gF5Mv>E5$PDd3VChH-^i2z0ko$2Q6(`GP?!s*n zSLQLphZ+EBycwn;cpmpLFPmc5+Tgi+0!PtF`yTCW2QOuNtaf~z89s@9j9l=PntHhn z(J~1r*!LhRDjhJ2_%MjYVh20BrH^WZ>H_RB%}^Ii+5PQ%%TkE*a_HuxkvkH^d-@JW zxFaL;ou4?m<{90E4g$zakZ&)+Qqqir%ER#W_=ZI6`u_Q!E`ZRhdKfCm_rS$c*U_Ox+P7>&HyPB0NP(Bs6$zLZ6&ZxKLTin#ZAoSagfIKCg{3V5 zl=uI9007-`K=v@~^=QIYEshnItue(zLrVyU*%m6FolkKJ@oUe5|4BvTiT|SoiAe*N z5aRbhU-zuHf4dDsST-nb!pa;Pkq8AY4TjN<^R zarqL_A5HCbvR?Vl^AuOrcP>yUUKnuTHL!s+7d%= z-AeTMqkWZ?{Q4!|J0bYlA!1zy*aLeP56Jw|Q_CGz)z(}KWkA_fl(IMQEnWTi`U!#9 z{r&y?uf{F)%$s8LK|j4C78%fCph>D-=#50RB;>&;dRx-$gT30$&Ylc~%JSvkTR5cm z*q{XfmC`5?hcKN+na?sP4?S>!l~@k|R9lcRt;^*ErJ%cPCPNFzKO-@0n@<7C>PJ(ZNM*r|pG0(SdJK^-<97!5r8;N%AHCb*E{0$vSA%8SE^)Ts%E{iPJG>2Juq=r82JAiJ5w^^$%HJEgQxC!^m# zCZ5Dy&b0JTV3e+kEuAU-`0?X*8EX9884)x~EL9h$bZv^aiSwWlQA3$_#l#ai>`l^3 z=~bRt<@1k7(}AAq4cc?$m-}!b;3C179DKYCY58}+U8N<7h?*{d_wxPz^Z?9Rb08-zngUBBz1b*wt?rFjR6GJQ3__-&t6&C*QZd zx~Ap`Y=`Y^nM+^3JhZ2OH$8#rPQ+r=9cOv<@?|oyt4Fd3(FX6%BDIQZRH8aPlJ+Sf zbamt@2R@uPLVpp$(9!!#D>ob_AbLE9A}m57fC~2y>>n~ZE-Dzz3nQT`Nb0pe|NN$} zug~5SdrnLkNGA012>J)4M8tM(#UHPx_rFw?0c#JB)*E<`%gR?5HvFtWM`XKED-A(_ z!sf_o_!AAu4lolF3|;AGagH;nH3@l_M&`qfL4Y8K(CcF#;c#^EGz(lQdJM-70OOIi z@GQ}7lgMk|M|5gKYKE%TBY8Lj8Q`2Bd6iS?&{OmTFI!sCMHAh;VFL~RL(;TNeK;TB zDf}6!3Y&7`;##r?~YyZx-t- zo?8XkwI~ayDIQsiv_}hO9vJ6K#e4!G_u$y<2{-%j4{;H~^C282Q4v*KwHp%a@7s3o zPQ&U4DAzMKLaZo(pwmL7uc03J|4kZ}ecApWb?$#QnD{5lXIMQ_(M8~h)Qw%{I!w0C z-Mt8GLR5;AH+%!c%)8?A;pKXgj5`iNw1J zX+A1N@F>V%q)^g`bCba{V#-1)jZSg;fx_iYm_^uW^Ziy=F5NkRWrl63IH}Gt+;#25 zIx-CTdX(%&hXL`;xvByJWDEjR7L#l#*Nig#7(`50Ln-LFJ6uw_v{}p~Fkxt-idJ%i z?Iq=aaJePa$;=B}9xoF!N||0>UPsWB6S4?fb#uFkLYloi#q} ziJ1*XD`(|yYC8er@Z&N2ka&W*iIW=4-|KiRK_h9A{8>1R7vk(sPsouBSlqH{ByLRo;oz0YBv7JY?A3Ichl_f3*99gPzm=m$R;iu3+l^znA;@X(qZVA0cR0U!y=qgEdWAk0J40C4mOGyx%hrbi| zE7lqY-aX~B%?hC53FfTow>ml(4(JQwj_~}8u*A>8AX=(^N*LZ=MMO3qil2FaaT1o2 zbQ{R*Aze@KS)G8NN%01lEt8ZQmsv#AGrF7jaUO_o7HY-`ycdf3zHPtxes0S$G(>+` zH+FpNV?(5tvgFhaAc`au%zD;!l+V|Qq)B||?CcM_p7c0xZYlW@qho zL8nPT7jFq45zSsEWdTbOWky z1qQZY-lwh5!7$&Rp-To_+%3vMwNKhE!E{1sAhgqITvDEvC?Ko@QeXseUY>$(NW)u9BCP;;3DP+*LjRIwlFU%p zAj;Yj6?bpiBYGLLQSwd_aKFhBX%vShs@tA&tzHlV6%bz~%BIq>w!{=TH;LijBlra5 z`_M2Vcx1@ZlT(8INuR6uHm%@;kr!G#Xw9zf%}N+!%Z!GApg?3K(h#G*E6^IjB4gP; z!8P9!qAZ;PTi^&W$xVsvO6uuipUksv;{!-CgpKk$jRJj==itHawm|GS8uJ9~ATz=u z=3Nkna8%bzT0Az6L>Zae;q}858CiSO`0h}!1Zu!@y8f9=oH4@AjbPiKhH<{E;O__p@i-jHFgY2|}Fq z4-A+@SAbG5!9f%;3-beBt48Lt`w)qufM|f$T4It%Cj6?jAum}IBA)rGx_TGtP(j)P znLz~j%BfWo1qL#p+*h~5yF3Y?J^l)KuLF*8`bc^TA`Jb zckKmCyt7A4mDaqcLsH6>Nb0bo%Fzf0WSY#w@=&f=j#x##yf{#EuY)Kr@Or8>kF!0h znD4fEBM}Ggr(xpH+TC^vPb7uocj$og1bSQgw&v%**eh8XuO;7}r_)el2qfVSNnCFVSfI8Q?ret!QCDUL{HC!B{af$P%(@A3{z3J9K40fXfh@LcE2U#I#ygvNsj zBVL(BKmmezAWe**GbQz(Xm(^U2{Lx`k*u<(Af*VN;Mf`GvcwPsH+31@+AI*WH5%nd zLT0y-(7(G+41MMvr4mcexm#Oz!SF-;w4^XYNmVKjIswhx6E@yCpb*_Ug4RIx z5mIvKmP_5XYoxIXc0I{*mee!Moe)C;NTZ8H03fET3DW?tHy^oWf)n5}q5s0l>Lxng zDA9!@Y5iqsX*cGAg>hW~p2xBkuX&v1K!ZdV=du?(FLzMBMAP@~+-VMDM!hAVvkF)! zrX*qaG*0orICF|yr(;AZth8a{4!)0{8~;McbUgrP)D>*%jY8D^taZY4^QKMty*d_T zK)#Ajr(e8y5hBuY)LGMuLTHYS3MK0}ZOLols|Fr}u#O92BO*OfnuXxU8v3j>j}ZeT z%TlUCsTERBcvO_FyO-B)(p^`Lkd#U;G(cwN0Q?mu)3P5SMg^OaPlr;O-42>5h1k(; z_L^Xj8zk2*S`bwC1q$IfS`B-(nC9rQ#+Nd{clZnt&`Jk&Vva{* zj15=M2%?JOtB&taM|uK)$$}Q-?5(wD_g8@vI}U>ncSU}nJXd^J9I}ikDM+D046N$t z=&)=_ypul$1x+r;FFUUV$D`B=Xy5ruYY%cUi=4=^o;8&!vCRacUl18$4;R2sM+OJ~ zf$5^I#T&)QzA(iE;Rx)yfB(LSYfTV`$wm1YrFn^teq^eXECch~-s8B7_%4ndA`_;& zF5paW0(mq59L4c06Wyji2snMIs^eqroQ~^y7k+cvEP0ePXrL&PbOey)i2a#@6+OdGrCye&o{U!2xyP zGy*@yVPwDt6B)#J4J`MRkbQ|0cpDx$1eGe5BM*K9&~77$g@P2V0$UwmslK6buZ_@% zx%>>G4X)&IoyyU211}&HygV}PZ zPMWB;uoU2#?7G8D?FN8+Xz`$bFVen;Q&UTe8c9iLL&!1!{2$RDXXTRYoZue^3nu2x zn|=Vymdlr)4B^W_AD)iaL|;uz24P?npNP>5CAO_%5qRWCg2&6-jp|NVOQ6$^U$v_O zg6e9WnVD%3Zjy`xrB-jW2lqS7j#EBU2=!#U!ph33nPdDa-6tq_kCft0u$Fi@OY|?@ zm420EqtFU67a9Q*y5^kBje`LCFvu8+f%1S!DzWX%G3?;CdG-78pOtp4Gw<}jDlm64%wYBgIrI^XBF1?6_e=W)QXGljF zO95qXkt&c3R9%X`(li75v(W=ywdza*grEizrXt)2WJ(b$;qvM2{>X4=iMj~s`eqrz z#kJ*!>Ws!GF|anBbAjN3ln!-q&W$5Q5Uv*2tMog@1WnL+tN^ch3!E+bVpL00e8j+R zKLZ6YtX>Zh#biWyB1}~#M0Gh-baW#esF|pL;@Lj)?^EZ7uT& z*ZH(=x=vFswAaT~O!~lVN#br}$)rs{`=mw2<>1XE&PXVmZNSXq2DUM~Eaq#l1q9lo z39)(m`1kHoY8LztnHcL#Lldkdu(4<_1P%VfAYO2i6ToZ4e^s!CRf6alIQ^p-#metp zYv)$nygJcZ;w45sIM{NVH@^f{sodUM2jwu4oVQZR!(qoRdwWJgUfT~c6uAJ1Tuwz~ zNka1DLMU8AfaiT^?KOhItNj?XCR19xzSoj5kH$@`0?=+$FtWH|`{6&LW59*t6V~*X zVOhdvE0=L$iVRX^Gd4{%Y64p2=g*%{(1{L;f;bBfi3z2Wz%OE`TKFl{rF7LfUnaN5 zsQ{wvO@IIHOG`_Wl{|85H}1SdEhXMAOgY-U0gWh?(TR-%_-`fiudqfz-PS?U2?r$N zP9bQ%w@{%Op&tD~!s)2Vh+aWIwrOnZZxYlA7f}Z!9T<%+^|4ay`H@j+a0&86CgG{2 zGr(#_+INgC^q|!^0x+-(Qa(UOGDM=+7_K2^G$h_fMm|PR1U^Eihy2$mmEm&;V^KJT z3fM715RE176?J?*QL{))%VZS%doDI_*oEOoof>@y99D?IDM&QBtBsaGVVbaq@Gw^P zFGI#rLHv<~<-0vVum4rjwT%D_hikvG88!m-5Ecc}xT1hCKYydRX&vL9w^<%Wgmn!| zaP31T>WE$vZm_yYYZ2hf{~K3Dp&2I*6RCwBXRziEcAc<%3_8iea*RgkMC4q!GtYaX z>e{v^z~)S51d-?t$yqnY-(smEB2GHB7E0PGu3X`j&iJD^u<`01NV&>DGk}LVE;1V? zE~3XC#9SqC?HPCuGyoSlS1h@H{WW-EWy&5=FA0y=0sdl`xBh?rPJ1t+B^YH(c}n4w19ot&J}nc+d= z#`9jJR%w)bx{~MHvSkagsuKSXa%1WCCI%Ubf4NQB5_@_(D2R->+b*5~t4IWYJOxpk z;Fij>ql-kc`4Sca$6C4LD%z-2M5iTeczm#{xL%}_(RoBK5iR~THeb%7*Fgq%F@l=! zQPPX~We_k(kcFx~sheK25b7xp6Tf6ek{X4`QUg_dGKU$20--BW4=yhm2%E6s=u%oS znj3PzlpiVzaS#I`5dYF)5CxbgkaMm;?j|G*na0GQ8o0R%F4}!<7Z~-U%?W`dBGzv^ zm<(Bf3=fcrPc-oK}Pt;XbD=rD!!0T zt-}2!%IJ!Cc+XaRL~>TgM`?#`bqA)&Fi$F$u2p>1Y#eTD@>#H@+YzicZrpG=M%?%$ zFz<3nrQ$FpL$M_sJt6MVUwlG0qTDgO-Xh|b)X}4tXsyr86v%ZM(8n(W-k<#AIRz8C z+-C>U)Kh2#f@kRD%3>uo;Ma!v{wx{LBA`dlq1D4d%yD8U}u&;)8b{1j}z`gk0N4gX5U+wn`~_!Z)&4B{9- zhf8j^AZtfH-gseOSW?oVzN&t=@8tsi4~XkoV9qQ)#l5A%rWr!A5aeiUq?{65yanAZ z-+js$x0VdX9Bt8w^_h=AxMmEL_AdXXae-KYi5yIa6eS{kE=wM@9(cu%d8$PTy(s_dXYt0-7Au@_S<1%td zG*`_|isLFcxisWv$ThnDc^1WIGtv6HRTXKoaEhxFWf%33G-}oesS&P84H_y}0`SUV z=wi4o>@6k@$YmVh+^nD!^JiR;uYXDIfWlQd%VgMk4&t*7yd%>PIpH6E0H-9&y(uz6 zPC9#dkB~Z)iE9;n%y!I7g7u|~#|OE1?MB-@02!x+;A7yLayUeJd)bfsEV_Us6tD6_ zel$%d$QZXj5v7~NHt9KWf^_pN@0XP~+zYybiA7X>MZgb|bFmR4tzMnPzX-5FEUM(4 z0q)w$y{Pc%cqN+1$)T~$!}V7ut(wn~TU`jSh0Q^~!dnL=q`xfmQe=9V5 z?ryVAF$4l)nl$suDKqkLX`Kyb9S}qsfZdRP;t;(c!&QhfJ+)~Rz_JZ~Wiq$`5TI*# zS@-)?{S-WR%?K+EJlg|)eO;QW-4fj7;!YoKAl|v5kw;p=7SjqHiP*$;U_b+wgR7!p zF^EQt1mK!N^&I2#SQ?$ej%bs4UnnHIq`iyC&70u!O7k}TKv`%9^+UW2Sb`IsER-2gS7Nl~wmcIF^W zB%>E`yETH)A3*^?tZ4!h?dg4Dk;n+T&G>-L{%-60_Dvl|U)(uE&?o7pOmJ~oMj@_v z8pH*P$V<+=E8xONNa7tekoz>s+^4o1&XeH+B07qj%e#);A0tgG{~$K zGEus+Q8=&K!8SK|m}iJA$8S)UjFDebJcv#T)+zzXoEMb{ejv8qLWfu`uQ5RoT959k zlrWUJVAlsk(t&mwCW-ucH{5z+m<9xff!T=gY%z0AaRA5g{UUautnRE2tE5 z*#Qv}Fwt`{TjmcGp#;!dQ=h~jAz$iOB6r#D+u=1omcweI3gZju-(Z+-{0Qw68urHT zgRrOh>|qK30Y$Jmxer1?K%Lx&`}uusF!vH%5#(GFyDMZ>qF@le^n@Gv8SJPHatU#n z!M*~laBsU|4}S8>*}bfk)9vT*`?V!UvD)5vxovoX?!+a?gCwiv#ixA(ZpaLUA-zB)=E2DQoTSL|01+AGvIY&_fI*;6Rce@rg7rYtoAB zo(#x43P%aiCtoFt;OCDnVppCWB)|6r)~%7x#HYu2Q8;(`^qV8lHqA=toftzQg$l65 z3IK)}x41{hMMU^Kvq1S)-~k9+P3!-Wa&7gV^d8E!^jr9yvhZO#dMxZNV+T&(1Mazz zmYVJy1%ilYVdmZ%06TVvNCEBh&M4 zfoFfz`KgBW%0E*xe>~po=kCP1CL+-J*UYDzo7kfQC+_)tTwRt# zw3wI8j&OrPOimR|g1)Xlw8ARGPg63zy}h;v9=^y~qviO`)NZ!_{hptN*oODjbas z!_m}GcvS!lQUKFPjNtwt(&RB8Xc?%0ow~9Kx5M=O`M-qBLwHFXIzaH*zL^+%3y#Gj z(uz1<;cL>DK!sI zu~v{Z3PAm*4I}FkWe^yx9qkKoO(Opt_FMUu>HiCbvgg0o=Xtq1`(4lOb(hE}M3JGUb&bW_fKq`*qkYa!tS=Z+M}0(xpDvl~EtZ+Tk3SL`ju8v= zzkZ$8{2yMcefno4%T{a$o^I?jxl$bY<_~h=_UWHt$-0TYJp*!p*f-c_@JsSxnJ{kj zeqKJe92yZ(FxJb6X~;AC*QMP(o5}suj(T7?;hHEsQpoa(78+`L>e&AM|MZ`zI_?7Q zC!j8;9;~rzG4gid$|~NM07ZxZ4Z8oj9Q-tSYsa$?^a@GIM<0exmkbJ#(I;0=m}7x< z=Ab^%8^F*pmxY?7N70AqG)La2?wHaBdCI z%R@@@C<$b0%^KGT5u>eFHJ)Atobqz-UNv8X-|k zrXnfQNe72H$uMFb6(;4Bhzb>@oI{dCI!GQuNoP7p68n2UteIx^TAw{@?{~j@zx%U& z-p{+NrJm=1`2BwOeP8!=UH2wcgAGRGbI@?f{G?w;3gu0Ic_@Wm(yn;|KGD*=OsCRj zF~Awz4YCQ*&rs>PwL7M|4z)`Ys4~=UU#6;`!9yeH-wKFdoHq^7CC)$a4#Y}uU z9DO!~q%hg^9!efTfmI=;4lkNNbqv%3N7WfDs#zJCT-WSit`Btj}SDCGyqQ!e;HEgTu~@+P{6 zN>k6v`_9Yv8+dCupJi8lOrlSxH8=)>W0E&-@CI3%a2ffW$#7(I66t++Iebxd%RTzP ze)K^{ZdQHVrHeL5KAUn*nsg_HMIqE?k$S4DJE`{@ml(2DtQv939JVj0qQdF!Ko*V# zbwL&&fYZSBR8{1RH)3b2KD*$p3rle$iqGU3$w@=P9~Z>w4t6rb#>Pf?ia~JeA4tUy zjwTBYX6qzj5Kmr0Cqu|LB&)Xu7FRyGX|C9`sRxwA;25Dz7>!bW!}7york)Up{eU>Q z2xQEM?}f~MfGkR6WuiF_PUHv{wwcIfxY&t2;G^4K(in|klZYl%G)W;NT%|9ai5sbu z==<)|ISCO)JFC)Cl1Id%(Ao@XAq_~*>7x36TWj&_cisz-A#xn}{L$<|mCyw!QdTG{ zDL=c4f*EHvh&lUtUilN#2>-H<$u%7SN7;mH#M6+=UPQT-<1KXu`J?-QdV`a6fK@Hu zdw!bzh~tvfa3L7r*t%Mrq>IUMzyV(L+u2qx{m9!Z-h|Q0N}H185uKEpjhC8^Y`m@6*w;ypyVrmI(pZB3 zFtd|0QRU>PB6(!N>JD6k!_fp(4`$&5HD5n;b=5;JVpWUGw8KXqZFcx<;x^r}#Al!1 zf9TFOXpseX7S-HaH*wC>`@8OKJGNN$`>Pw#*<&LC3NoJM6+}kx>8GCpV=oHvp!>Z7 zevTf1WqD5+8^n!$i@@$Hw>2<&j*Rin>5e$; z@W)ak`2yiNlU|cBa4w{zI8yfuN+o9KA(d43Y*YGM5;wub0>pca^5tx}dfMY5;#>)s zfUo>c+8W~%L-rpcZ^mA)zgW=Gh7ltm*Lk_5b^Nhay|SfdM0zngkho5j0TMD6PQekd z#DS*L-9SIIa7YFqK{O?cK&Kv_VXxpw2n!FEaBiIdiEC^Yl5w|!xvM3Hy|R^P#&|5T z_(1hlh99U7%4wv>EWTc*0wanjhHhrCzR)5uX^^s%!yf`P?Q=)04m?bscAC5VZ11yJ zVuZ1a%K!ypd*AF;LMC$^;AK_2+%~U2ef;ZBXCFAdyvilBYZf#$f(J-VATgVsfi~3* zqFj+fvbq=N&IJZr{Bk6Zkm|(F0_b)p413#xUsWHux@8w=wyWpJq0^8L4$c{vbK39y zZoN5>$=oK%!B`zoIp%bYv%8rOuFtos9`1jAIcM5S=5a}Z-!G4UQJ&UKZu=mWrT)-W zgn**r;(Y)Kc>(h{w^;+J*Ji2W?3wS^%b*FfeqZhWK}aN69gbIsVpg^?p6C$`Mn*>2 z6jx2K+w_vJ34wMpd9(Z(ts3$0J}A)}&0{SqomDei_-U}TL!7?>vkV zL?a{~vw#+O`qGx_o&HoydrGLGNZsVnmnxlGok72n+T8^?z^l1YpQ^rX`Rdn61=a%6 z_z28J(Sm9qMmvr40;0@(^PGKCH22+#6_JN9#K8q$ z{<%tQN77Eyd?u8WxD|~a!PNl1;;bPqVCEGA=AM{EgH*KIi!U<;Wk0;0{ZwwzYsg$H zNB=RezJv1s`Uz{QFALBlk$!8Nhc9HmdsiMWhF5ps^ZW!DMtI~zlJ6fB)jeL)yD z;$O*EO(zYTj_}nY89%O9n!lHJl^6DEH4+zV+35W^TW%$~AFM9U7-Z_r`9?WAR4Q%O z@J!)MJ1wC8l(#QL6H<>f#1tx?w6NO6oIprPlpn`G)KOD=_UaML^h>|W|6e*v=j5|m zV8zY9)u!;G>M%AiRghQ(*c+^|8LvQ&6RDmw7NF9pl(&{ERqC5sb*0sW7;b(}<}3XI zgQ57-hfiK4HX5LEj;v+5rlW@76JqoMj)R#|`2;|Pg0{aGQVE({)~p{vyw!>To%Fad zz%EEXLK`luSgCSfW`Zlx_b7zR%#!Peqk-^fRFtOtzqfAfI_s49IiO$xM{zjzWH37E zxQ0Om1N_w!zI)Kei3Ub}q|@`I3F}-tU$WZ%edUIcFYYxDmUolmHePvpPp4NNgGG9o za%#l}KP)TjZwkQbdLW8~ybS4=xDnMtcAJ$(isCN(>F6J{OP+qQlq7mfzSpL)CMF9! z{&s1%DmL#3Ao>)99-XA=|MR&KZESU}eo_CW=F0r!EI(JB#Mc$k@zZ`bn%gRpttZ4Z zHa40mkp5JJTV5XEKJ6m{5}2I>4_3TG#4?mvK?V-T8l6-91v}yXEitd8Xzwzq) zhoZY=(;4G6-?WS_c$R8adv9sS;=!Aa4{xbCxAtD;kFM8WX9|)qWIoKCsSQm~OH^eN z-O}UWKICVarUOHYUxSeTh2@(b2cmQUEJA=wF)+Gu1lQBsH%y*ty!(feyzTR)wjEv< zgCl((B`ZJrLos+|Z>JrkSJsL+2=_?>IL1b7{#77Bg%|@RIRG;qjo?JFGWmVeHM(f0 z-l1bCcH01<3KrPO~kR zviIJ6_P-ElEh(mEQr?OM3R%`dKs&_jqOvZ(Gwvl=4d$elLJkWO{eUFp!go>i%}cXn zWs@|7Q+=5#AO}_c_e;DT%ukE1(4nTHA5dxAs%ad9**Zu>(}pFR!~qw$b?ZX&mmKnC zI3q&;*|PXD8z+Kx+H_%{lysK7aPNyTL4MZ1CxsVhj7eNtex_shQQH8XqF)0c1<4yo z4wwG6)K3rl9J2e6wMS20!;1-%;$KHZ~ zH#;@X7HRpw-=O9Y7qQ@KkotwBk+e~^+;9>b8UW%O(lbTUm@Ldu3u9utXLC3$VY_V= zYE_uX2ZZi~zABJ3Sel2R9uas~s3@?Dip5ioOr-;Iu-BGPV}#v7O)6a;i)2a^&t1Qv zm}gEV1lW!u(GQ;-BIaq571N;&&=Pk6In#xO3^hku!Vbr>$>yL~A6WJ3r9c~00d^tB zn37|0?oXfElu&#B204zL5Ma{BJjxXTqgJgvNlL6J%o_BF93)4q?(nJ-%AcNklGf10 zH?IW_20^y|L#CljJ;m%6(`lTZ72(oBREQ5xSl(4C6D?D&EJzE`-b}PE@RFp67ptv* z@x;1Rxp%%8 z_t#R1Ir>Y@`mE|>whbZejf(T<&o89d$(eVzmjm8psvHImURd672tcVeUYzYPY?dxv z`s)$Hp9w(VL{T{Oc=;r`jqmaP+_^d@-#c6J?tAYoE5B{hF7wGp-Me2o+Gq21zvLOO zPk!I!*Zq<@T^(DWGIH6f@}B(1gkZBdf~P3LXz?C^(XDqw?3qlb5&M3vVBFMrMV1(8 z`CBWbR9OMlS#M+XHZV4qDVljcAHemE-B3BS-9~vSmO7E$4eC?haeHyw>jXIe|0t=@ z@?it_GKHr*6YyklxaHj4^T{hn#rImKJa}t@&Rc4m@-%WrzG&1QGQ{AA0f#)}7Ys3Y zS8e%zqe$7|l~?v2X-Q0$Y(f~G5S8PDcHr|wb`a#&Xz?U24m1&prR>kB(0{9 zH_9Qky?HasrDXC`F>eEt75=vdUWG>335)RhjY|U9QJ(u{D3oGJb7$476^@pSP7u2+ zNwuWU$N{r>Zo+*sKk}O!fm4TMeeco9D!%wrU9R}dil>L4-UBmmS2e#hx0oEzgi9_s zFp!+FDpLW(p2tf#c5oCJDrw;@w@St5i*F8w$D|qpL&X?X#KyUX373#T{^>c@4iFdL zlcThOo0VSV@dKtzX}rErFk=XxNz2|?z$0}}6hh#mM+>F@P2a`4;O;o!diy_}fARZa zJi_$ZVSn-y<27OWCjrr6-OL~NBZp;X^HOVR$%TFYhysRJ1X+0f=YM$yCyj#S2Y*@# z(u(>2p2q)p-2d)q)Bjo-mDhsl#&4SS()^;KBUX_*N8@2Wsyb`ZIrs8eFze3+Ys)P@ zSFoa`n`?qS%Mop+eeMBV1d2NC7T*QB8Kib~cm|4E+%1H|`A_fzuY|4fcf3_|emJSw z>!M7QVNO!6V`FZ6Ae{OT=oZbgq7>i?DjF?QLU?TOQ?#A=SuPhLvrw|@zzw<@ zcomL#Vpa&Z9iZD*fTa#ji8eJr}gxlqscAC*N!M3jJGho0K)o>kQ&VayeU6ku@ zV>{8n`P0unyZ-QWf*Rx}Up!+y)7mY;!m03;K{q6?+ zvLYerL9;JTe)Glkb#;MSWQu9dvweL}Xlaw1I2?;zdT)$K45YtkXq3M3>sa2aUlGj~ zYB5F)gy)~yyk(2op7MkcqBMv@e-_N0J38tgj(>+9?2{hqqxdQCkHTlEU)J6moE9@^ zxS!3(#k28?i-6cTV7vL-$TG17K}t1BF} z90#CGg-ZAU)+#4ABF2dqO7GqK#bsmpwx7>hTTX7fWJ#6J^I1L5A9*L&aCAyWNJ{IJ zDGw_W%r9A@}0{hDpE2W&Eoo4%sOUk47S8gK-2TsVDjTFO}6 zV)3U8z9_8=+%OLUj3~lZuU3-HW8T2lMfz?V5jvZec|)Zd!)SNZZ0*i5T%>iN{@^DkFz%2Ch|2w;Q*)W~8U4UbpsC?!}$MHy$3~ zE@GiTWWp1#Of1vE5F=)1;@{LBkG8p?mukDrb5N_vSJqeDR{rSQbIm(yQ-`+)Y=7KF zXQS<~Hpb8&W(fZeO8|o7Wav@S6`DCS$k^!xt)x;crbO_;WT!b4#iBl#&a_>j_%iA_ z+Zl$L)ml>j4wKKrSTkIo)-$gUYCLf!piiC%)51JSKWWX9|LQU$$wfcBTf* zVk54hbX>B;;Q)8n*(;ehAkGJQd3g|3673Ddg_UDh`Xr9qQf)ICps)`RLwn#r>DRgJ zJxD`AR~|;IKJ%&KLM0bbY3AsS#LUTI%6zbXG;GQf=rszKK@ckVP=4;mk3NH;m+>HT zdpT)fpq40Roh|ttEmLwfm&7I*AWbw@yJE8exBQUhX_6$bAZd4a-tpS^@@3lTYuT$22H}i*)g+zt?#3n26{XwZG)%%TZalEMm@|`oc zF7p5uSe|NqRkPspr^Z{XHB8pu^4e<9K6%iSo1H6nt@YdTDR&|#zU!JyT(E&PW zwz^(-)_%CatMuvSZl86(8spn3XvNl=mW7=fca`aE+%vYq@Jh(qd!g#dMjf9#)(rLb zjkk*TE_FBg>aI!mvnvZ5p7}?o@11f1+hm0sc1KFD;t;(we`bW6r+g(%HE@5ovZ(le z_;0}%aof9oX79}r1B_>tKZ3M#WR88qP=n#aZ{B+fZ9EH$L{VcfJ+SZLGtfE~%X!jwyX?~C z8a#HGGXYw7XU}qQaLB1f#~}6wikDuA4M%a=Q#Bf)OMxS%Rp?4nGh|Sj8Q&0O!BE?oDEc zt$KZe40CKVulnT#PtQ-LY&6PDQ(AMuYFicnJ2ur*XY{8NZH6Dzksu2gx+H*cKKs99 zg9Qc#?v0Eb;Fw2QgNTa@B3Ye z9QGb|pi>;tgVsjV-S+MA`+?qF$LKc~uGPW9PBcu3YnOR;Qrq&|KdGtx ztFMu%-~Yx{HUDv256^$V`UUp!g-wB#m19CRe8RRQB^5mB{5!p<88%;_^?TAXycq~T zxmoJ-pZMdkNu2g-tItK170&YX+;wfe*O8DUmZw_5%zp&xa+lEeOqxCJ{|(RaFD!=U zkNrRB#~*&TNM?u1yyk$-oA(fq52Ax|cH)%|U-Nqe9@62`_@@FzU>G9)oT9!X!E$8U zBk9J#3LaKUvwhZfi%AUK$~h<%6=8g4+G0J?(#Qm6HY)bXd%NA@fLhvy*t+`MPkhhn z1uhO+jQ|gT--A-2?m#rke8X%grK*~V!d|$8%nQ1|<|yc1cVjYxol8GupA~q)FGx9m zzAvFciW-`ykwu&tH^eIK%9V+@p(qctzE^+KEMXBigZ)n{*1gO=K9bn32qRUw^G|-z zHAYi#8n=9>_`HF!0VFYx)*QWoje%iWJi{*q!i89pyEwVMX{vUMO#gke3|9rk#KuDQ z7`QpLu(Gm)192+P(=c7B)E=BZs@~nRQ}y(KV>^1azJmts^3|(3XU)4*R!;u5-=&Iu z=)h*vuK4z-=LjO~QQNa~6t+sg^|RY7?KEYBQSLjhgID?YE&MuHS{V%UX;bX|Tr>2g zx>u=t$_B%h1qRF7-o2h)r#_;akOr`1G-CpCg~j=djEwNB8x|H%xMCkm6+h$aw+>BP zz3SVz2ikN(2AYbtJlJPiZO0;~eRDVM2=+0r?Rd1J4`|PaDH@(csu^3VlFzFZwrsRH z`?cl?ulC7Ct#fa@Dln+_^)(6IShfH0GVN4ny?NeVqYs9oV( zJ2b1;JGG4dy3u9Oe0nA$+az>qEJ$D9xzh4xCnL?fzZ5@EeOB&6QZ4G+O1SXt2P%t>>;)^6-}l0ZR$9NExo6TZ?H%#kkiLX% z|02fx>tkimd1AJm?lx&ld3v@OwkdD1NY9kY4gDufnX-i}8}zTVGP*!NpO8ASx9Bt- z2Co0X@rarCO^HV9|}E>W`g2$6xr^$IBrz2Jjp3 zE^jM478;Xa)?$)&yrKz8)FBugb7MnOZL1%ZJ*8uMd9hwnQa20Qhl9~(@qvAHV`5@# zb+>)afLKR-qW$=L9EaH3K08j_a*K;R4EcHIR0E%t^R_>nk zbKm_^uGT6FcQhe-Q3EdXwBRUq?k7Z*$&>v-dx|QfF_9@_wAD44I4t5&rL}gPa23VT z4MVf3=u>Vh)jQ1B%3Hq{hG7H?TzXItJ`M@f-@bt@$RdbbC46?MXrE(T)5O2#cw1Up z3W>(~bdv`s$^=Mdxy)icfVwglaJ4}23hhK&a{LJc+ha~5fQllF6V(IrhtEX3WJpX%jT6*Jov4Es#P^$? z71`TlesItxJr_GD{L>M46Qs1E^-KTia%$?4gIjV>JYq70%*lj%J`n28e%epMR053d zZTjnH->hL^(m>Nh^s7-R5hU|gM<%I7=Of3Y&$kEdUIg7isqmB`UvLt{R&J(<7Ef%3 z%Er+M2=+!y7p1`j2b3O{Fyxlg1G4K9&+X5k1&MJ%9_<@{vwbW^4Q2ZOBJsY-{f}@+0{}zEH(%Ij)Z#wt4U3|w^ty(3PN_*Xp z&H@P)JAZyZEhS~&=v(jJ?$y7)nT!w+e*l)^L8yuR&W=gyX*JgdtOBx^-g@2LS1)tJ z?+|c5UsgC?N%1d68gd0}yb_4|dUcaS4MoQxvt#AzI9==LRFok>&RwyhR|yYOciTj- zxxu{kDS*#-wl!SfSZ=3p|Z~$QE3GWrPVw`L7w8QNy*xWN8Fb=P?0+Zi7^&RrF<{!XY2pp9G6(17whl{Yxrk~+^yfsu zz0fI2a6EK55NbEMj>gC%o=ZHIoH8y!WyLsUhsS|cQloNxu@mtIj>!z)q_w-j~H z@d{ySE0+w+=+SqbYm~t{EuZ#}`F;y5utl>C(Q7|2vA{oEz~LN*zpX=)^^)nz(s}!bXu;{#*MqU z!Kx{EF6V^U;$od3obHsa$oj((ZWyjFvBW7pJXH&6J19$J)t1~=d( z)MTIj`R6u9wbzVA+{AnzN7pfKtiGGuBj=U?$9Q{Gv_Q}=F?qLNd65d;je%@-)1rLsuEwa$hg<{;|~Gm9^=^f%}0lbp-m zyXwrz`GKWY;&IYo93b(Z%<5;v8f zJIZpB$0=-jzKBlFLMKP4!KF0oWTqzUkFwNNq_R;dYC8{)00T!9KtS>+5t22I*4TpW z#yDp7s?k@c7_H~@OtsY&H!d-RlYPrMgPJ8P!e={obcKf7E)^g9}j|%8iLw06Mb_6BR*Dx|bO%q?X)m5vRPU4@6ET zqJ581JM}*uGDP08N(jT;$ePhv-1P{z0c~5gigEOW7|;0@fID2EZ@u;)t@4yMRtNC8Br%tD80p?{gJ%GE_2CCuBAMWzFWUXk)u?9jaG##NNAEsq0;)vwW z2KMfKF+3fyKeGBM-02qf^+!8?EjBeXlR+AyFIoYkLuBP*Q^EZWuYJxFFGZIoLrs_s zpj~nEi19|l>b3S)w`-fqVyg2k6d^=qe2cmVU+uErc@5g|5Unm)1|QMRxO8bS-*1a9 zyx~`;pHX{1uCI@Oc^adEaX^%Zp8qP`4*c_gL-dNk%5+ImJ_FSBjcie!Mn2h@{%+kX zXe2|my1)dBiNMmcE!J`?Up({8HZwEhiwGk_5Wz1rk0%0-gGBO6NXUu{FD>VeBA7pb z&wCN)oo}Wenyb7CCTp$4x~;P!y1|npCI5so`)>{ORrhW$xJNuc0ksc3DLLDPCfer{ zw&*x%_34uz(5~~~Gh8HGotzk$9Z3&llHLYtxVd#k(0HB3>&m#*Bs;I8B@4T#`9Pid zj7;mvDn?M~6-giQlhkO#cz7bl3UO8y7gUU=hiDafZI2&6Y0^-bp&K|*@&@`myXUt7 z>n!~vz0YhoiU~Uguw2*dG}0ZrAMj~-_*B|hqcEkUq%P|Z-2mrRmdRC9~1*OkNTqWmI;sTfjqR{Wfh zkl^V1-RcFFOvd^ob-BvJnmWfbd9hwMlBq=dZp6cJ@Rg%p)z{Y#8~1Z(m+ag&U3AZj z6P61~^WbWhSMYhfgs^tOx~3Ng#^0X*g6qzS;b;}dzs}I8Cn;%l zcZTHa7FyLml~Dt`T}ZI}o*ySQ=Yf%gI0~Dqy@p2R^jF80+;twW%YkNJ^PDnc505RU zCV){TTs*NnKP%SPKTKt_bWe*qifcjn+4i z?S38Sr+qaAlhZJ1QPlJDp^~Ulhxc?E0HzYP^W5llA36McLtd?0!`5nYVs5 zehm39?vi!MxsPk`Vhz0ym_5@{WBF)84o&hcEWmKg(O7geQ6?+0u;QUub0Bd9)4Wn+ z(}IHqebhAg7EYHLW4MbpDE^C+a%~2>^a5zzq}5b5?^TTUlC1r|AlY`Zu3Tueo}^Mp zN_uvz0*Oh+dsH+2#LKStW>bjO*>s8QEEcf0^?HtpuB`{(m_5%GeGYZSY#F?h|DMZ= z#fT{-0MndVXdc?Am)YL|`v;k83p>aABnyqfCz}JzDu7Eb28lk?`rj)aV4TtyB${iOG5#i;3(+LeEt(h=*W}UK1-x_v> z_YFB@v1hK*xSf%$b?6T7kB?_mQNU&~sV%svW|6oY5S#b&2BZ~F*0Ut`1SIaZA>SkY zq)jt>UOlEo4xBn;7oFGKWtW7LeKz*rqc)`MJ7+`J#bzH1F|QR+s5>N zg@dEao~7Q9v(f%Q7nA5ux|i&*>)={h)1H+p?W5;;bNo|}$8VsxKEQSRM}})Tpve)_ zrOeDsG0IZXKk{&=4pi6wc;3hv>DbMPKbhDUVIJy|Fv`M}jSR9Nxo>rK%%qMZNrLpL zG9|0=E3?+AFGf$e0D{}?U0Z{|)@1mmXadN|>nI8?+v?_Jd2y7f`#XqM^^^|XEH;pM z_!{r{%H{|TI-*T>TGDWQCD)e3rrE;}j&SSqcb!jC=e~=Z=ZK6LO#0IF-P5>R@m)t_ zxRC4-oFl-dvnZo`rFsWn6;Uf$zK}+wq~bBjppz8)$xmVal}aKXoX-DY`f{vy1%;Uf zlET}okFFUzVZu5VjrH5p6W!A0E?c(jRH>zvLz2SQV%Q+GWmz0p*;2nmOuMkl-Q_$V z)Db}*`lb%Ef)f()7Ch+OE;oi8UMv}DJ^d=8A4tdVYmT&nH96=}^xI1}^-OsNMu(hC z%$>voK2V|94d7AcdnCx!k5qXPg#=~XSJYZ^%gnSLBhvG_6z{Zw^~{mlpMLy%wlc-T z0MqTGc1~q-cWRAc$~2D#udO}n21Lhg)KATqhm@3* zz}Er7Q50T_`_#WSW7)`M5^X~^4zNnN_B?N2SeRJ($tV{meed_ROob>SAHk%{S?med zXYRSxu&gd|zG3Xr_psi;vXxP@ia3BkBrg*-!U0XhOu*QS9jPOewmomV$P;~7aLZ~A z+7OjRG2=7!D<*z*ue7cail+?wengr`J8pyc2J&S$E4NKNY@#^%w@XbQbTr$>uB+>G zpWGw@t(x#w*)$GDc~_gtyv5L1Ol(Legdi{WuasNN={o;>)4SBHQYrFSr9(%Hh4mD9 zx{bd6x%XP^dEK|WTyyplo@tE#aJRTSMRs1H3`9=Mf9%Ps<+vt` zyJY{aDg7!fYyF0y+a}`EbkhD+K?0iCx$Zs*AMwg+_vTj?6?>X(+GNm<}~a%QIl zXPGmR-BeOtGuWiGNGHR3CMBca^E#PqK^-8PENOTJ84T&;L#;!rZ4dXKKY%H>;7pF7 zNJ%aGzQc%x#{8mOL)N~z3@BuUvgt45FQjOyb0bfaJ{9*XZa!kA%HfgCsp1&IeTw`& zXZhAfZ*O||xek+F*&P3<*T8}4&Uu0Foh#HZDvvy|@>wG!yFj=Ibfv}CKc{8Cc_q$j z3mO-U$k8zH7?4rZK5ea%59>!GjECjnR9I{i%wx@DF?`Nsso(|otgmhxA%sh#g^?f;&eRYLNBl{%5@Z!(R*OcC27A`Js(IB+iCIl5v}Iu zWYaIAZg$kt(z>avuH!TquMqq%d9Y~A-u4<~l?hj41tHsG)HV3qz2{Ysi;AC1S0nE@ znQh})HMh)Egr~i%yN+mG@Z5l50qN7NGDkJKRcfADwqw+QImgP^RXoRv6ZVG~Ig&WZ zsNV63ZmoU!d(0Gt$)H@R7e%%ag-va z+_Ssy-sm%n^@_d9P8R9;q)iXBkIgHpf7?0ic4NMXKFNJp;J3mLgDBV&Z$kOAgD zT9x^pU#1h_lYbLqrqEFnt&eDPv2L*;xFN6^vXo!hi- zoi-#~yrCh~U9DkVjfLSaz56D1fnHI=$(38A9p#A1;a5hvOUGC={U7F5S&ZR;8yuW2 zM*NK8t`&3v-xgXT%*1po<|tBU$Pb z#8}_Ix5k?h?FdU&Fq=WJ6$*)?)bc<2`E^>}<9xOV1;v6AIN^{OIFgYG3SAczw|4DX z-|SeiSl3|4Bi9^S_{j#(d-iw0dwACCLN!V=y(VLy*5mm`3ow9*jL8|z*Or(ycJKu9& z?NfQ&ggB5%)e#0aWCh8JOu4LM?w>Zpb`@?U+@jtP3|J0SD432|^LupC;~zq0$-_F7 z#99C?`egw@LE-p)m?_9s_EGW75icZ>NUeIsXoK8&GU)uyihAqt(9yf+v!IMLF@$mytYu?wAFi>y z;4M&u5+EEhtSosf)UDG1myX`C8_6s}jAIBYlW&oV1BQs{V?Q08$zn+xWgcd=1P+7P z5Q`oE-hKP}qj&~h5MOZXML6g^e_okX)m>v~f9u4aS814{{{bw|B+2U7zWrAP87mi( zol;Bg!e8MF;!!4XF@TdDz{8PYd}P-FVP>Y60S-k0yzZx;z5=6UgS;7HFzjwsHw;7} zC7IyRlMT2N7!u-S{qV~6^4v6;nkkMg5V3OPL?ADOwCWsX`81VuODs`MGpU1jQ-Jo@)6=^<%CMsy{fX7IcLYW!t;kXL zyuAzkahY_7fkPLUiaiLK&{R^M=k-;G7KD0@lLBhe9M+}kMM98~H$)OG0x&K$vQG=3 zWt^UJrS~AyR!OYPc_DS>u^9&lmtG^Y+T?B|yM$jO!|WOJBV*KgOU^Ef z3mq%}?a-6DVjUUXeUAstcj^1%Hha{E0eT2C@CXQR0GK&>&skI8? z#J9h)br*^P@nCvBx`Y#QPFVn^vh#eO{3zoGNm@(%4JxnU>qT$u!{B2wBTDQ8Nvx!8 z!`RaU9Ga!z0J%&Nh^%|BP3N5)G?EE}gIUN+VbXwLdz8OPHf3!)gb^p5@?ypc%}rbu z$hf56qe2SzWJy%wvt=p=*~IkID_6*p$EEUl&#ikXD#WM~r7U}e?%?~IiY zA)q&*m=|Q<4>F!O83vCyMI=AKt5@n&+G8lA(bV(bLsna!cpE#{7#qH@2yaG6>w+k^SCR;Rkz z;lL|>5r8c9&}CxL3fC|b{ghZSQ)@j6n&W&OK*XxlZ0;Qf=Ss%O_{&6E%@jqwjF8gN zC(Y$>A5@ZQohNf{+Xvs?d1?HVDHbgCp#X^_U1ya_TB!*Rhzj=ub%eTwWgTPrh5TQlS9`|b5_*ce+{ z@E+slK6Zj@zmcu2m5t!>2+7T7w{7=1c=6(y!nOPZZ|>e* zlcT@N8zvxs%xpQgd?|+7u&#ktcyw)%tqPKP0HKDZM@w83XSCJIn6!xi? z0=nD`1)FdI6w2*;nlJV(|BaH#>!bJ2-y-*y{PP!?v$Pez7Ltav>5bDKkwXZbG_#sA`iK9)7CpcFd>9M_>Yq!{`I21bAQaArH#mCA+# z`qW#ua=rEnyqTXFd#jmsL(Y?pgM+$yBa?4%aPSe8xSpCT3eR?4zcAEY?AQ@!%73bb z^Tx-P^!yXd-9AUXR{IF-C^*4jxO&Cu(kz>9+aDhv$RzIEyO--pA#MF;{lxPRH^k`V zyEeYNc8_*^A@TjJRE5aYi)SpQ&kG)Ie5(*4wRXb>`r^V|$nMez=>VnZO9!@ds>m3Y zzpyspiYZygeQ?UAL-+=rFF1H;yTA{NCY?ZUo8lcszZLHl=lPy6*KbXby{;+vwYS}AnHXyP|ihSrIitEoUL41c z$(=oWc64MUNn_|IZ7Pzt;k`@F7e`#uJigYzIi#vGm zV8`HK{O#MfWzw7{2h_CkUA5Sxe0_p78Uswuox4l^|IeRqaA(0!cW`rauidz@E=jkD zm1=EmO&8l#wcF|DP_uwm){TpWkGH2Z9hdN;ioXzYkbAh1Ien-p-mNk-Gc)=|=a-m- zJv(lBgsuJg^XE23Mvf}y(G9Yp=zMjXaSutC?y}jLT zcHFqIs3>ZB;aTef``gg0hR~M5Qbz z#Ts()nl(fRi^#m^Oi{qbBEdSUh+TV`)8)7cVRciKK1dr zn$h|#Ou}=hC1($=pFWGNAs!~~=@GBn*w^0AidQyn_LY;X5G)p|n8 z_wcp&_P#!oRI`Sv>AB&2MPU_2fkSwEpWnO8d2D5u)9abepZ@tI=JUYdpn1GQPvwF0 zw9R|>nrY_V{C0D;H){Ry!a$Ur- zxuwNZ)Ma)TmTXKqBO{}+@MOKN`1@NkapwGH^_73FTUTo3fg5|W<0wzI!?4Bl^z@Y` ztZDAP+P8*ZL&e$>TvfbW#Kr1T^@>Jh_#K8c!{kB)`K|qq>$L6?bIpC~>#G-^VLzyT zo^}~8K&f5WejxB#(iH|VF|qd9iGKdl`n?(v27uQ!GJ@P=+{i_^(73&5DM}OvhJi)W* zn4pw}5clU1G1dq*9 z$$-nJfr?{%UN3bS^OjCKF_)Rv$rwzYN@>$|e>!41LT|Uu`EB)vD|6=S7yohl{gwjlR7RZ%Ec(@8IBI8gy;f=Qg(` z^WRsm-c$Hw=P6dk_wV0}V{wyJk!tkqW23f`lJDiWDo6XXwY9ayV-zA7cJADHp7vDC zZE>5>HY~3K_w~{#VYhOy-s_{UB)vIMVr65K@_7$FpeA0UPW||dbiQqG6}>ztARyp! zvL3ac*&oYYdc^zPgLOL_=dIOKjV^~-{rE_CtBX-&FT8)rq)M8{_lWYB#-}4ABit*0 z`z^i+Wt!i9sM%OV{_!@B>)Vbf;XYIowPVuN@|^7O&QVjs%o>vE6{Q6Q1y^$H&E2n` zWGMJGHkRV*O783K-Mb8@PMtDkz|Fcm=GN<~HypQ-mA2n2UvR=>qv-U9Y+2rig2w0azvL0{c z3rwY7(*o6L*Kmd6X!7e-RaM(_hToS~imq_@jmB_hEgqyKkNbvwCYtYW zHCVJ*<}c8vu?akjizS0cC;IF5ppx>lGUm^ZO2-bqpC2v#j4z;QdECGMJ1HseT(flY zoNBYn$9t>SQ^nIvYCXvEEZw?gi%0&_V$;yZ7_7O7=^}+>2`0x+jjh zy=>>R1x63>^Yd31v8GV}z^9_$Xk-LE?>O$#+m(17ACNP#GxXOlE7PKA;1T}*5Mjrv z!w*VR z&8|}x&F}otFAGr?k2Bx_hDkm>$RFyW;_V_Xx?F9VlXM?WY)t%7@K7T$KQr^R?)>zK z``fo~QU1vy=x1G;6|?&OVOzV5+_~miQ)`uqoVlT#3Lm{(d?dY|p}nh{*JajL7r1~? z&>s6b8$a0-<{hVxVko{iKjV2w_W3UC2;RnYbJG;Lw<_^iDW5&}#Oud|NHjQ>MY1Kf z{HK+|OIFIxqv|z*Vs7tSBzG4S&o8jEG?jZ*96PEeQj~mS$$DJ$d0j$~$B?B#McCID z!yIgkUF9LBH*enD)!N$nc{qRR<&T9GLw~R`-rg1-pod2z>)wc}`Ph2A=dy0wdza;Q zQ6W0_^_Y;WX^x1~_&ZY;Ztib_43-PC6ZFwXTiJKhJTlr`h0Z61`{|v&(eW24wb(fQ zm&Tgdyp=Ltim!=#m8@pIb(IvBd#g91XShF|k*h?<81R4S?v9^7q8wZA%f6IUn`Tm2W`Eji#m9NgavxXU0N79qA`U(mQ*$Rsepn(VcJg%)KGE5Wd%s3>Te7!bS8GU?l zZaBj~{z}rv=QmL~AN=U-)E~Uh&c6TFty><`(>MDC4cHiYY+A$reM`@u>J@(#Ik^0m z+7?}&Klf;M<4GPIm>#I2)!J}zPKCaZ<2gO|yyVNGL5J{{uQRUBv>#87zdxa=sp)aO z@Nt9ZhTpF4avT=9b>`SFEa<3gE^hA2?M2?I*nJ)S{clG*OW&H*#>T3r#cJOgxkycx z+hCjXklD39;cH>xMLv@nA6uK1;dw$5z$9tuW(~CKdyMxztJ8Ltotql2JJojUS?8|mpY82o zXzn?q>S-p_WDyHms!8n;szNZo`M~7F_rLFDTerJ^@#P2`>#pdZ9%+v(tdCXZ%o)Xt zmZbI7B__Ul^-3uqRK%HX5mb2nGIrp=mnYje-Xv({NYVH$ngd@?%e^@L8c^j+Tbs_E zJ9lC=vyvxrTT+dmh5J@CePwb^33L>G=Kam*z6gIiN$9-44btJ;mZdLM>{h3 zn;ZX*R!fOIB=;g3ojlgGF5!lQgDO6*Hk`WEZm^NBQ0}Y=&Cp*b$3C7s2!ob&JCB85 zdD;54tAPU@-#~m`k7M7dUbnfqW}AwLrb)EjINk?ByLF|6g@sx5hf?y>$~rnTq+h0q z@bK`o)XTjTjzPn!Q_mnkS|R)|JNp5YZcaf#!NM{Al%iHH1Ez|O`{zx%U2iyQYQ7*~ zFhn`evh6+IfuF(Gz}GKDoTX?9I{67YZVSrCjvYf=2#^)R)@pT`(%L6*|BvrT12{#&S zw#Nsb&ki(G79v!aI(#)57uFn8?^XYzOaCCAqsS2m|wdUnacr#Jell2}6 z+V$zF??p+Y10%e~H=DM9_H0bne-bGZ^aj;k{lI|(vkOygik=fHC5gFQ;JD8d8-0CE zR-;Il9HmWVfY0}S9vm8axfMJgj4Hd7XUkH3lJ0g!#u#isNr%)U$3x00T#W0~le=KlbucX-YRA+_Ddk8MO)E?chG) zo0m9M;wn)v(~PSR0(RpU z)RFbOPx_#)o;!bjqSbl0Vg8_F#uqQ=2SK5RP9=d~TUow(45}FK%NfXC>hPPpr7|p* z<(J~+tWr+aKER8@SXbEup2AzKR6hu8hHDP%w3W6QgyySv-)=+yYQk~fpcN&PR|P(h^$9kbh3J?JeZ z=@se#aFp)IKJUJMkL}XJR6m3coSg4K#d&>=j` zgc6&(Q%EPT?@q T6NA@egt~9o3Bm3J04^tihGXtEYWw^j>MT)rN<1NNVw*hnH8n z#ty8aUH z$rS*Mr=Xv8V_~jsuR>%-z{`%Ti{ zS5(Mv6dCtvu}^ktjj(6Ja#VtrLz>g;^AD@)3zlM36JM^~ek43zBg3RM*O8_-W~3}B z`PY`MTPNrmmK;#U@PEpGF|$G~8f(lu$O{3$Y7p1ki#ozTu@&D3k{*?l_qCaYXX}4F z-=?P($2i;6)cpMByu*q_6Sa65AD!atz6_?n@y$Ik6oz;5`hS(C$Up-;6| z4DI&NqsY``TI;HD?KrBId;!Q)5sRgxt1HrNaYjP-vj=seE;{b!U|QX_9XldXMBbwK z6R`^oh3r5V6ggV>_wU~s(v(42Ja_X+5kM$t+w-b5&$<8a)y!o5#*aywHD_LiO8E^O zAnsYWehr}1tC991@*OqMi+EDKusK(*Txko9h&)jX{P-_hA+Q~Of`pne6Ws_hex5vYhN|m+qSJ+!)o8E3Z=#m5K#%+jRR1_ zWp?~2J{%1>8hxL@1&D01fce>jCylEv?BO@7MC%QUjy_zQprrwEK|RyzH6Sc2Dk&C! z&=wne*90{_qsslB%wpQmyt0Si&y9cHV@Y%;R8Qac(?8e6Yvrgw({KW=AEcXCg+Dr? znh>`zH)YyW`C3X!s!G>e#9%B?DLL+?gMO*MGU!xoy153_A8|AbSgKZ2s51vV7i;YN6)ZIjVYHV_aIzp59Ek z!6e{hxd0g7(b*Z1(ds09R6VuwiyvqF*6S-BMA-9NRu=nJIX~z`cT>=s%IbcZ@_w?H zzC=JH=;r>yI+4h_p)hYz6GC2Y)~&0DwX#)r~5xt@`d?Ou4D znkGxQUGY*XN8cwpT& zX>Ih{J2(Ch13KvuoESnH+2f)QEiEm9e5NX~+<8ljbNmyQE*BdinzcgrJ6W5JPXGDS zqpgZ9+ERyP0U~}!nX4w?&13UE%1-#R*FIB89;k+%SPgn6`L~6(vw-CI9@o*XDjS%IM;nTVZtUpj z7?C}ss^b#z$NJqpMa*s>7g0*@uAT*YmV`Q9iv_j~`o$^gC8t5)uycR^{Wt%}tzUny z?nB=~H;qEO>cqY&jdnPX$(_0;7 ziiVhN_2cf^9bC~+*kgcjoI0LL3>$q0MyWr1hdj_rDwylV3l|PU z+y-Z(cj7TALYA2r%MJmmZ?TqEiB;k-Z%m~EY}XN}3N5Ky0+m9$6F6>l{l(~#8H35RfDJ-n%>gsCJ_CD{W+oEHzuAhp> ziWMsepuxj@o4+_K4T^jTRjD>YdM6|U`NX`*=PlN@9raODS&?ixUfnKJEm2SfdioRd z>A-stxT4a!I@R@i_-laFm6EO(D(UvMJ!dO8q4eg$8a(F1wb^O{!(Pl{36N(@ zad(1F-^2Z#x!Z-B%pBseIJ0NjrvnQWRq>@#%uHM%Hl93 zO;+Q}tYN^sdI?1R1GSl!-Ngd_Xa_7)-#p?%gTbFa1^bO1=IwS?FKRE`spAXz>a#a< zq9Mcq@F<(--NII<-a&#pE%Ybl%;$Zgx2_Q2V`X)s;?Bz3kIF;E64CSRhFi5z$gS;9 zQ+(LDM_6DB?KegrCKd~d`uBlLl>HSyt=Hh|XHGG*H)dFdrW8=|}3gyHqmj4y6>HnB%_5Xj+e{bylf9e?pSAT?Vz+j4jabeJJ zPoX$GY|1bC;(zq*j-%>5P(KLeO}ETm0JX&?5qj)phb*X`T4>nr-xMI)8O1X??mNRbB~F#=M}hR34S0{o~a z&%?Ap+OuiXDdkIxFqRe#_2b->SJ z1hEhc_c@7cUY8)R->Qf1ZWGF|5Ev$7f%skV#rjyV;bXqju*t*gGfYF(;zyEyDGvD)mQr}OO> zo$PMs-bW!AeF51eMASvFa0yJhd*By^QtS(4p`Kr~Vejcz9Ey>}O-&l5{zt3v#TWau zWARkO(WXi(Dvq~g!~eXD?!#wX#hRlJ$4&_(n#rl>wb@@h<`eF0)eLuUN zSyU51SOJzqZJZjf3p_M?JIj_XkS26kcE6(FzB46kfy(E^559Qj#@ zRua9f&aF2hVB5BB;kY|JL2U}o~drn!@AI-Tr@}_K49%M$W zQj?Z!yWKo`e@(2LhFZ{=>zG2|JvL)6OpUYzUNluWh&`Y?8UBTzK1G9i@VmWQ!n0P- zcg-V)IKbZ*i{NY5t~m{5bp!qD#^*bZ8DPQbc#F;&MNVcxnjjZXqzs2v$D*!M0pJhF z!-}Ac+ji{=?7fM^!WtO7KCJ!kZuPo(uNRk)XsXS}HiIo3@4b7VEnf$`IVI@)_g_B> z)U$14z|@F_W6HdrpPwEn=JWJ--1}0(r%#_`*cf5;CBVHfhQ_iRE*`5f(Ngl9GeqZq zosU;b`D_vLbzp1|A5q;>lww$MWdHvC=SSGR>%Whz+HVdrHX7K(JCd>J0Qjz$&W*(< zGEt)l89jJ%uqg@&hJL6yCJ=3DG}<^|-j#fgso*}gF*_51z4#pst^$#>;Q*}rYyMP z`lU-ga-pI}-Q3)AiZ+P|3s;%DA&>yV6^Z&7{3CF*yCM^+>j7phSXqF zMn#otWZ^fL+*$~?<6Xg)QVQuZ$qS{&^TX>ivZh*{W!rZ6wCpOObZ}=F{!A(OZKLoj zVgSt!L%;8XmrF-BgwRw}G-=wGfa8ha1HE`Tb%5m^?d>eo(aA|Vu2nzjN}M@?(E49| zFNX-)ZQH%uaNe14=jAY8sTk z;DGV~G}K~{_x1BS%5635Z-aeW#Jt90Ee3%DOiMs<@Q>gZ5fNddvPgI-;n{=-vn&kd zFS%kXmf)Kp3nyRUq8DxAJ%XhPy|)f8hIe1xhdaMb0HpQQ>9{FlQ%(Sk_W7x5m-tJp z+}ltDVb^p4*QX(&uTB*RUoA`cjqY3^E<<1|3n?ymx7N^f+D@NPx zGEr$?Mft#0;=Q3YRxwJ>30yC^*#O?mM8QU}XzTXRggw~8Rg85U>pG&Hn^`ucuM(%7 zrw;S2q$X@j3An@?;Apb7vTw~KBj8b!@8-dl$tBbn4CgorMTS8ro~EF+6x zM_Qt99DKCr{N~SP)gWucdHXa!p4sKIll1t&RGtwIm(6TTf}vGoeldANBlOKHO3<|x zjs{qWP6e7lk9GoBS8y@lj*xMYGG`x_S5TnII*Pk__5l2MRfHxIF&G@bu0m`n4kc{@ z)_A10%1?5i5P)k@7LFC1umZ?jh!I>5e-~6h=;9^x`P)x{URyV_x@O)oVxhx?7eZQe|@NxF>=#^T$~HLcyBs7 zsxb%HLOw^+OZI(1?VV`iSlI6?Ji=ndj;L4j57^5GsgM8q>m7&plPA?!{VDpv2u>*B zUO;JIp{c&T^!nTl8=Gp#F}0c2?R%Yd3+`Jd@S}DsIt*;8M1h^ zDIZ!mW(K!ZmHn7bF(@N=g%9H5#9))E?DLaPS!ZTvLD+8DxAoFeQH*nVK=dEnzkisO z^#TpHIU9mnK?Ock@Qf{um~X-01UgQggrEMj_qK^r*)Jw z$tWvd(3r7rQ~e>3A)29$idqT%2^YR$ZXl)Nq-Cpc@C32oT_F4rEhufE=UW&-X=R+u zE-o%pdDjXlfA$iPPV z`HS~eNtIK<=i~YF*A+A5Lq&NQ3IwS{_U*WkOQGngAWg-mSGmDpC)cHc92mjF2uPV7x?oZH>JCWnF;sh!sY5lR zt6quH!@?aMA5Zo2DvpI-tH!P`BQHKBz!%g~^i}F5D^JAIj$@jEulW_6b1q!H`U*Hi z_R03*HRG@D5r%*;JB&xMT1S$zs8x3;oH}(y;G1g`zfshC0r z3OAb#@hILhv3Lz*OLvW)Ju@$-BdyqDTw1{f06?kfs>R_Bz8&`-dfpK}FthG!K>Q0qSR6jWdX%u@UWs$a$a+?LRS=gWkShpV^SK;0x8$+{Q+MwPmYH7c_`uEB1tCB;i z;SnfKUqZ#)vXdxT|K?MBXJ`8pC*b^X96WgU)B%d*_J4A#y+kjBNB8oz&h(1ry#Gk7 z{yz%Cf1PiYvl^)9ZgE>$Xax38IiJ-&4snP@+{4O*0f^`lMEZ#_w3THy#imOp>hmf} zu?R!Kr9|yK=*X;5pRED@QR$4(0_8jlf)H&a9xUpk5TBPrn~N%-M)IOFYSUVDW?1?m zySDo{T9j{~WbX0YE1-Y`AwdgiX4spvDDEC~_?h13>@3M9xUO1V8&7RY@bX__aWJs5S5KJoIRqg$IT zWE5V13cDG>-DsdtO{i!URp~_gTJhT-MhGL2UT)@pQ9{BL)gH|3VRDVVy)>tN3Q)2U zzJv!*;y=mThss$?5;jZl)M8=OMdI#SYpMG?L=ffJJBhw(1dn94GeAcVn60zKEifCV zG73;3Yk@C^Vye@G31CzVf?@onv^x0`5YLHX7H=c0=3aWJkV69CfFg(;Ar(Oh!Tz|= zw=m0k={BEIjjKCgS;jXNKL&KvX1?|7j_BL~Z}0{-%J-pXK0dEN75gF4rOar1Jb5BL zk#h?GYZjTBlVN1AKSN_*${l<+OlX$$aUuYoz+aFp%AYscoeZ=*$_cx(kfCG2#o+wC9Wun zWi;@wJWSHri8`G$+>KKIah>;?cz$`fkSMUxKF}nG=!BG$7R{$i^T+krQN`h8|)Ih(GKcFZ#8>iCrD_jYrqoXQ%<8L?E1UhwOwcQLCd~${6azGYAPdTgUMzffQLTH zChgw?I`SNM?0-z_c^^Mo({x>vOA!OZ5DtF%9_XsE&>fd2sI6jIehINki29 zAihX;Gr76WE(7O1fXktX=Z*ta*MCyOCwMaQia$Fft6*dlP5e!8bAsTGwW7I@PXl39 z0*WMy9LqXd+;g|Rox@Ob3^<1h2{qtXRnI?&ah@8Qg)IvEK6#r&EQK<(0i@j*$xE_4 zevSf0)w{H5gDDYVn`}RLszi4779RZzYQ0{@BGK)kTl5m$9-H7KcrO(w_ZdQ(JX8e` zxN&TN#yAH6N#a1|*bgc|gFb2X{R|*LWx@i{v_U3SZr;qq{nfzo!F^~DAA&A$^Vh4F zj$MrF_X#TDeSf`_lcAAjqWnCmE!z``2vuZ`D)B38l5{0tkV~Sx1A>ND9{0OJUE%LUtWq4w*Y(uTKn_*$ zmb=_9FUKf1Z{P0Mo6V60)!_g~;S$iVA@uESoGM>4(Xc&NGu`}<9fzKUpsqMPdPU6) zP+D=A$Ytpyk&4CK=!@$46~6Fl5Bzjf+6a8V`kEHxfJyQ*#|gsxXE>fFii9t_;|68( z0C*UQ&5j6Ho>Q9a0+s`UmXK_soZ=u9rel|fUH$v6ICR88f20A+|H<6imlB8eJdGQ@ z1;Y+u0CeKLRS|>lS0mf)34tmMx*RU_GvRde8_et9y}q`~JxRZG$9dXFqiJG;%TiWO z&K~sXG-J1#=*t}G4q#KefuX=Q%ulf~db$A7iDTu&l}dQ*x~7@$+H#n9VEx|AVyy;3 zKzTf@jLg{0{_UZnF4;f2x(qVn@o*TSOr^x&;a|#e!u#gSMi}>|vvbqFhtTP8XO)$e zw%&|Bdwym9?sJc*1XQV!og*4#J?%WES()oN#!AJ0sZ7={#hU$Ko52mq@?S=P&r=2f z^ygB0L$X;Td;uosNcC6S4^l)ZJEL8`@_m8(*3bV`vvhTV|7VFmJy5m@SR6QH(X z@XPyPJXR4E2F8>$4ak%jSVVYJ&%K%SeqmArn_Gg&iSl9#zC)sH8N>+hhm%Dr;siYC z;?h!SxTngtwyEL4-r=?|Igq`mA&D5WT45qq2@U76pzD%eCm7U@e~~VWygd|P?dQCP zUXgMku}A|(3MT2}8E?Zt2eIr98&E!kbi?PwQ&Y%xxufym;X{qKTbD2vK&^tCv<Ln8>pd2Dq!-RHNMNXTuOv%@PO1V9+7Mw5_SDaSuU*iP$d9fJn@zrnJJVH zTGxQJ5U2!Oey8CtOnZ%_tvFO952QYJk#rnUNF;eo=yAsTvaE1L1$A|GH>ZC*G;cJ5 zJ$V@&l?582dirwKeuU(z&3R{b&Y@Tq{pIbQp{A1W>Jo;?8r^Mbu&K^R@9T5E(n#$K z7w-S)?>C*EomFoymB}$KksJWJ&tP0I-sMgvEi|)jODsZ=D8r3G&oWh0XJJGj%cepE z7xVGs$AR=yH-BA4xF1mq^4w#hUSi`*`*S75WY~1e6<}DV@F|0Y*WPQatfSky^e8tf zvD(=g*aGJ*Y=8;V5*lmd{cB@QR4 zvyqFCFU=1HjsN-ufOwCr*M_tZZ7A0+s-$MA83Hy#pp$cOiwI6MA-qp3%iG)Ab$&!* z0oX>+rpp7N6E-SpJ*-tD_eS(BQl-tm&?M|HT)ar7K|MY~RZlky!cCg>9g?R9&L{Li zPLQUxglnN4R+H@vddXXOn$SLcg`tpvk~yXqe%y)}7a215@89o%U>NYEwCqkXY-(Vu z05~ti1R=$TR583$z8f8q!84nA-e_c4*pKZcwFQ5J9g>TwkTmBh3&)!`d$9I1VpMZ) zT8m(wViq?*WCAq_4ZYNs4f_OPgO_LLLeAp1;f&~hD#4T#-2bJvRyiptiHNeKiV|E7 zo=G1b^O$7D`-6Db+12I$Q_>5KPnw1gA-u{23;c*?<~ycazs?3bp*kw$IZdP*Rxsu8 z)4H$y_2^N_rLVVBAb8!lgy+jvBG@5%l;AXo*%P0(%9Ft{Mkl3n$)PVU(**4YqH%lu zD2^G@Jgn?&TsxkSd8T-EJ`ymfw|BCG2B?!GciEsd{1R;MoMgnu0eq^4ychKt7e2}w zS4DJexPtx}9=ugaYDtbF#Y-I(Q|IyWjIQ37){{?1CL0ri+pxYPCo#}kt^?B@qM%AgycX?p5ta^ z8N}lXr892|UveojP?B1B2&pz2Dk7bEi-n=e(%;Z<01gyURAUpF{mkYz#!PHg;FND& zJ-e}(6EsmE8qAj#4plNjqprGIQC@T=oM>anDhZBTbf5*6l$Djua?ZyqNJ)hs%FDLQ zoACc*Ba@k;3WD7IOBq+VcoP-p>tHgZ_+zjhI0HATp(y_!h~fIaj-JA>I=wF6B53)6oi*vHXJdfiqI* zC2{Bm>RT6SW?hb7BnLPqm_-+-b>a3IL--+)Fvw`E7$^ZAk;C9iX79p77(d3ClDc7M zCz~WU+Eaon1^(ATBcqkM9%5y^B0+5(!up{t)_MPaLkQy;8>C$ItI%<30NtUZ&vXU% zRYg3R*w_?zg5f+79q|hM-%s;~Vd0YiR%~frUfy;lCa9ze1DnEGXI#F6Z(jmyVr^-T z60qqyj4%f_o~`h&rMlPWR}*I~e_=8h_>()hooU~`c%)8ZfVXIRE`#+LG$H!2x{xh8 zNG-9~%$5kj1a=(I4p~LXbjFyCB)HfK^DX&kvptX4Usmq1%zcgEWe#;Sn;NsmJTS!G zMita>cmlDoA^PBM(NKl3o{zgfco2hpDt+_%pr?&*g(KpMm@r`on&pC$P8aJ+L_C-z zFRDi(qOg|K-`~n2vLc9EMHlxssE9^TnCRh`A)k{^#}ybs^Bw*5>k>AMw#$Sv^noLV z5;8!hmee;(w}tOv#_}F9f59Y-VL~y+>L3Km;LRh+4+MpfuWD79#Cqta24%Mm^{2yw z&}m{IH@qdgNh3WD+t+n&@Dv2u$ZWaMKfOR3WANmUU?2rQ6bYL1kanB{s3BL}Yi|r> zdPx#ZLn!C*j-#pqq< ziRFHjYF}|}C1vOsZe@9)|Ko9ec%4p+F+kvcOcAP(Ec3h<2a?@{j$uie0`ZewO`KG; zb{@qhw0(csSoCA;oir=rZ^;C5&^iv9cPWzuSii ze?4*a1BKYss5{l5QA$}i^oeea1V41}OR!*CZ?>IuaI;de5-S?i0BZ&?a&95 zV-yOVxGd8kE+{8xszGMq5wYJd^lPp3{)YtA!fq5q2S`uN2WiKc?jhWVA{Fbe8xH;Xrl9H8JXQ|f;{=jm8ETUWmZ9Lu%YbXv2jo8vK7^^_CY=MnPrAKx%!*mX?QXR?v z>N%@323Ht(nI%{HBmiU+A-S(Q${Vuo(mN6O1Df4)tnU4t*4!Jo~bl}L+QXPi1N!v!^ z_lRPoM+>*iBnlA0$w-5~Ak(ZO+Q5Y=azBYP0`h*zZGPk-iAX(QSxw1&COq2{)(C0| z>Zk8=|AeThw%fu)0J#l7LiPglJg`rq41zRRoKpQ5Pb4JdJ_*Z)FjAX0)%uoGHK7MZ zOa)I{2?RBG7F)CuMg~y@a!Qy$^&q)ep(Mnj64;k`#-*F843v-o36vdqumVN=;|KW* zehE>3=ucN*=)DdJaitgqHZraJdO8t89hrN_^Oo5*xOe0k!r3rwR^PrV*|Jz8L=i(4%9*-bMbfxFjCleL8btr zFyej2Go(poR|IZ^mlbb?cNLFJ09m!bYALCU;D3Bj%*a?hIky0eOvYPZZk;$;ZlnW_QnEz6N2Y}4~TZ{?ulpr?es01KMc0F>0{a{ig zSO?7TA|jpaGn9AEg9pzc(TEuip<&O}l>GtLr%Ni4NFe!wNHa(V@IkdA8wDIk!vThk zr_{+*wGqm6EQVQGP@Zq^O-C3L3sD-kobFH7Av)d{TMbl?*Q7SKw3<7DN2a_;#l3Zv z)(h!Yy@QeE$EBu#QbgKga&x(&yhsF^S`W0HZ$$A&Sb zPk)F*Ysn2EVEtMMf5%XO@_v0>_gus|9m*Qp#*G_YAk>k8lt-6w9)-kdVcIo_0HpG> z_9AXvja|emLc0br0{*cVX^VjNrC4jUZJ7MW_BDZ2H4C2#{fSfJ9D-c}hw-F<;-QLH zqq;)D{Kn;}cN21S9B=PNsA1}h5CnPw11_QVp%1wQ?Gr^3Y2`x+px<442|1)z;akjQ zPe>||+3u9CMd11kma8e0ju$|4$i@oPh~ou9_$6_G^-RL`25PvD?V#BpP5lT^Z!pJq z2NwKby~l|D&q77M=V?na6fJ2QA}qmS+&G4#8+{<~`t|Er9Q&Z9wi%FlGE{WZ7ho_8 zah#)2tT;~@uHH|DYZSn`op}Cg=wz}XsjTd*77o}%q$gk%+V7X5P<*PZ`dB0xhuJG7 z5YII1c+7epi;A@bVP^}q8){}|X8ujyj~;zP^}>)&P$xfylDQQ%{%ga5aE^GDq0FeJq5~208pm6L!B=+}751=kPES=*qv|e{=JV zLXNvae|GOoUe`?z`R6lxC${#mGz({8OQ;tLroO`Sc(2=j?wpXQ!~bEint+U@h!47{_OI$ZGbVX)UmGb|~Jy zRSd&#ICEIIlmmVS~6SVcojBrAq3L3~)E`S6#bU=Vo|$C6wk$Wq}xh4Nh`>tIPG z_N#X^OdU*@=CC4~}&F*ymnUG7-(0{zYF9zA-rTl7{YH0#&E zRwYP1m0ZSmmE134!}~RqWtd0%q=gfTONJL%dlBGB3i_5M;msr^BZG;Tu{eEP;5X7w z5l@mp{P{9KOXtu~0>oPQ9NNT*M2>9V`t4Y8uOQJF0j|@+LA$6`ph^=sgM>;WBZ|N- z8vRR9$e#PM{JE!!Q{VTgCYl4o)v7_8 zGDhKImvE$x6+byor(-h)>0n9g{QEWRX4IOd$ghEespcDd#saI;*W4Mr4))p(<$Z*t6#?JT#@cEZ2qXiMD@e?cIqbureyaE#NxI$$RiISu0_7 ze$Zlv$%|L50)-5twwxd<@P0_LCzZ~Kl}m6)RpvNY;tlm+jisthVbdwYeTIq8MkT&G zuty{@<4}Ew2n6Y-7GMVSqt>7Wzu9Q0IcpiF!Z^oBk~nP~97A1#2l}=a(QPr6Iq`)v zvE#!xJeCt1mmf5v_1m25BldFhF_%WrB=5vQg17$=(AH;tHNHEPEaS zq#6sQfJ{BKBJ$z6i~Ag4kP5U|GWH47r47}T5KnV@3v9vni?gQS1E*`J$>1pPukrT} zcZd>HCWb4aH2j5gm%b-#fmi@X0Bf~8G8cjw!eAtj!dh}YX_=rxt~78l`vJUD4VV`n zw(H*+UyJ~M6x{LySvULyAO{Viqh5+9dT9yH-Z>VDCld`-3L?4>#z)AjAs7je_1&KJ zcrQM|1i3UTV?u$8N=9O39ns)Ov;yQ=t<-61C?4C|1pJ>CPBtqxF_{cQyoU525$s92 zz;@b_5QYI^*_)?t-nGkUX9ZEk-@7;?k75MNkmN}SL?xwz7WeEts!hkL+>cu*R;pCF zV=rJZVS3}8Kl_q3c9)& zMNZLqwC{b1$w)T-*# zqNv7@!B~ufkx?btL<1VC1Sp2hzNRJ?n9U7Ptq~W~D zMpGi=kpsUa0kT5u*1d|i?3c;F<3g|@CA>gQ2>@RN~?6${&ZTlaNWd4qN zKZfcT0CpdC%O%Y1rsBAunYlSm z1UzZs7|J^XiWN%06&QC9(uZ^iFg9jl*!?%QkYO8#2ppt9My-j$!y1D_6#5!cqR2Qp zR;nb8_KmEA?V9no2ENm2{NWgiGQljz`$A{=0?i-npQ3;zzWG_J08Q}Z7 z=JJ}2K%-FK%0MFyGqnmo7f=18w<~x*G43Fek+aS=Z{AE&6tJSqnBotEH_P5VOCz#> zR#p~KC+R#;dPU}c6*xhC#<@gzD=*xO*vR#xs5Z=#)3KnURU()hr;lbwCp`GwyA#l? zIiQ8&9topL(9SbSam5=MOt;1~hTkzQbsR*+-&tbAhYxHvMy{BW$%*;+MnCL=BL@zg zrP*Ro1fS*nT+l9dc5=`(ehlE>6wAQAe|BajqMl3}YdK2k%Sga|ljrv8XA)rrfuYyy%^X^-<2OMHrfs3IXt0ni()3<|Rt4_Mn~k7heB{ zaGJqUA*2aimlhnVfBudNAbV32#BCd}@kO93I;dtE>KF#(`oN0_kDmCr?zmpGCrAm& zUXV78bEwEsY{VgO!cigPNKEOEG2vo6gb5yb$i=wt=fEbOL4eZBkG8N4d|6Jn8C8Cm z9kzTwBTmW_vFgAzG_jao5yJt+qsBn)VNK(+aLAV=4@yKWvW}rx^$s+o%wpOC&CqhO z9INc~(?U3D4QKT-2-{JJbkA%pL$kni3me`j%xWQ_z0KE7=lf`#&hXhjzIl`>!EXQX zkLzyiy{79ww`GO9@E60KwGH^4o$E-Tsr z07>G0VZob!c?($+-3t#=`1~Lj#;mcot5FlozUbgkA(E*l&pPZiVdngZ(vqO#<0ns2 z{PUoo6TN5rpLdG@OFwkO8UhJ2xR9qM$xX~(8mSZm3a{9}&PZlMApgaZ#3BR`oJe>Q zLd9T9jw)%WNXmp*>;VM01cpjdkA$zI7x+ReBgrLPBRY{hX`IWPV<(o15AK3&Lc&*o zs}hh{F~Wi(TY{X31jSAs4OB8Cgj@G>530d4v{Ib+5C%J;2P^YA>LWRFfxve}S_Iw^ zJ}R<-jS&;*iP)J0loJi)%$eJ)z2x`-hbVC2?6{fIwP?tP%~Jb6dt4! zkGnQr;}mJ@q>kd6IwUjk!K;acK(4gOGCv+O%%$quAk^e>&i*Rc$m^?j1tvSIF{yYs zJt=J-&L?3LaiuGebQ3r^(hdxVQ&wPi$FKQ=iUSzGq3Ru{GS2=y#c-ZR=HyN99%w*` zGgL%Z?r=rk2;36!#C!rIgY^jxetPgC(N6(^^rDYaN%%$95s0Rr@XVWeG4F|A1jobDnk5kgrxqCq$>y` ziFK1hCy=crQwX3K%6fX?WRe~7H(QJS?o$>omQM{)hTuj{6>R#$;{DMM`qHYepJpBm ze9ZFWL-eJg%0k`czD6l;mHf@Q!3;8YLck9UN>#Khl5M~Oc?&rcrlwi8d8I_AjaB?_ zjvMRDTYYOq@1Cc+<5sCeS{d+X%YQ$LI>Bht|N0>l5n#tLrPdFumE;etMuwPjEZ3y6 z9nVQzyeQCI-oh#<*w}F3`-ac|T#f*A%Tj;)YUmz8{p`f)L7t=b_KM$9Em{PzC*)|$ zI^jMc{sPNnsr4@Rb03-RD|XJb81s*t+`Lzc2T^*e!s80XLji1cRA~f32wEaeBH9$r zz)V6!EDb!FY*)nX=sisWAVb zk9v%6C1)6ss1#BYB|j_KHDS3@t8hA|63&AQ z)?be0Jo5+hHzg$=&7Kdwc(6X;CFu|3m=j~nII4NoUQg0%@~IT^3l82>#L(bAwyxoU z4W~n9nzs+lV{$8DC!1Nsqj4BiBaQ|m=h10Rg4n%SpxIjqf=Jlx0hyJb4LZ?rT z=t9$HC8Hq88tH(T!Ggv!;C5^rg^z;4U(Em6Ac%vLi8PRK$V|p516bWeddx{DReg~Wy~L3jY^G@OMPcl)c&;bm=xfB|$%lFdzW9u(xbPWNj)I5qAWYp|*wYHBmdE|W7Dz)8*Dw?TY~#}{G{LhN09 zKKuf5>@CI_}ASAJu!^Gj7*1`cK4h^*m;|{fWbzFsk!eO>16}lfuDidN< z$MA54)uBMIe_qPenvEM(QTS{zMMTU8h_ux}gQ-Y)L7XxI=^=t>+9F9IAmw8ROAdnp z8U8r=@+^oO-f%FUWdz#(5e#yXK^-t@hz^M)i^L&5`zQ9EpwE<5ORLu2mv-&YHZfZe zK$-+b!Yb4k@ihB`E#D`-6+`?8ptqit5ldj10iwR-l87%@yhZ?-%R@UTzj2rJc?qJh z5lMNVK@rM-ve$*fh$LWY#Q51&pKx4YKVAn()SJ^7phhVmUN#O&VhS|I-w##x9L9un zCx-z_iZ?xc#(=^?`Ua?M1g?q`b8Wy<$|TzmCwA@Q@4*M2l`y%@TjPte#ESz3MG6~s z5-HFy`h<2L0t>tg2fhcgg*t*hB#DhY7HVC@w8|rtqZs7xaA?{cSx3lJX{edR>Rg&1 z^9EgIg-AjU0>Kdp0%@FB4ZrPm)jW9o>S4XE6ED^@JX(HH>(tIN9~l0jz8aXNDd)*~ zOrk6U7FI$R$e;gKBEt9lFCMS_zrl|FUvotdf>VB@T($h-(NMnOfCd9pv6S=V;(a^` zJ|}FBJRWr%9#y05Q+$b{}(*sSh`DBuU_?U#92+M$Q&b28ijrQ z76=RO9iPB;+@>->9DD$rGkOX;_G@G=)LV??JO#=N{E<$iAIxa>p#1T7LQN#u@PA-q z`qjTz5@nC1J$aDbCX-VrwyN$;#O)+upTUf_BODwh>e@J>y6lC3B3ArKKA;32KKx1& z5ytP6foeg=Ni~NM1R+D;$kQPWBSL7RUUUeqGY0>V6Q7Y0z$^kMOq+O}d~Hl}TA!4` zM-`Ca?II>;8F*bLE!k-#5)6~Z6g!{H;t=Kx4a+bZPt2(ooqhR{fiaVzLY&6^4cVDg z-J%W4YUvh$MB)JA0hBw1PBy> zoap)}@^-Aatna@wyNo|Yc5DJ8;HH=pqmPm>Jz z6@C#crG0CbIRr{Sq3uA&O_8^d!6kkaTEw_IXXNvlaQl&4lfMYAP`3#?idgC{{URR1`9_6Q?r zWm=s@15t#fZ{|;iqA{QM4T2%EnRiY;`yK8DQ5xZc*5acE5i>^Q+)!sjR$pg{f$Pn&fa)(#F8{~)&Xh5M;A1_%LUae)F5Gm*sA z4Sp5@5|R@Tdl&bgW&TnA_djxyqoT^(pI9JdTAPyNKQ69(6cHuooHma5-fE2w#Q?fn1M-I+(_yuW)td)s6Rd)pKx5k;jGQOUfc zWQu4oiw22OqLgtbQ<4-aWL7eiloGZ~QBhJ-Z5cu(V;N#|UKf7n?B|c?S?4_G{LWhE z+-tA3*Ftyq{r!GF*YLjH@AoBLlVI(PZ-nVZdafcjW_um^w+bo8G4by31=?AmrGrO2 zA5q%lboQ-VMwwr`P05cqrdnd=nNJQp22Zj_ndVSCJG*rip!X3ol*O66Z`;ym?iBLk zJLBP@bZ&V=`|_%)o8H||N1Qu1@JZwrPY;cQ`)&_UP41r&IK<0LrmoHO&+HbD(=_%A z+!Hd9$YW~$%y1MgN3vSIqOa0F`s;qRO_%{p4*N zJd5J?C@^=_aP7nkf3G03?VsO#vVOk#u72mDS1)X4@~CjY5G%vGW3+kr(9Z&3ImJvc zmFc8k?bFyMzjR}=KgG41O8yXajrkA0r_5E<{vtREJ4v^f{%*Y1n8({DYw4*hxzu%w zX#LvN>NzP7i70dQ1|r?73%G+%8_z37dZTyl-05(9`iipjqnL>| zvs4*5jHkbJFBKvk)^SX)V1dO^Uda)o+gp}a`1_RC!SdW9lPd9-I%h#MRO=o zs(^Wh2&V`$$Aqg!=_s&~DY^v#)!yy@LTxTqZ{REklC!zT4AKjzkwj8-7}VbQmw{<% zX>QEKLRPz_nSE%tZonqrg^LswmYbc&^dOEPWiOs?K#vLEJ93$v)1q*%wWI1m3M9jD zyb}P29fj{7D^neQ?zE6&y9MHhCw$Ls&w~|-ZblS?v!Gqdd<9>JqtF%{uD?isF#wK2 z+0oC0$%7auM`^Co6z!~h4{2c_-`QVp2Ka-~+F|5MAwewG@ES{U2>krPS<(?Q@`H6P zT5)~ranuvc`s-&a5-vSCCQ5Utw9AB5aTTJLGv3lK2mu|axQXQbPE*KDGz$bkLrFA$ zOTg6CC`3QIZlKOCK6B4>yI8}*)D_hfF;C1R#m9?*1CAkriK zGQX6T8_!JPmr0A9bb$hJ^h7-HL|j$oS6SB;E_*>jhfTFSg;8bO^{s&nc$tL!|E=3t zun?CSUghrN=VUKhr0Bb^K|n_yiK{bYU?{JD=4y|Bt%d%xJ3{{Vz0v(S6Nf^U&*Pv8 zT>~OR99%P{NreX0pb9`zn#B3fa#LZsXFbh zO9BKJkAur=@Aig#Qv@QSX^Ed^XA#4CwO*qhsSE^x$E^lsqqJ0F*T7t;4h+rpYN&?2 z+d_PYfcB*bP;0>eyq_gPGqmpeQ0U)2G;Zh^^SW?pF_|*R#F3Hiv;!smiLo-z5}d9> zXMR-(#YC&ttz$)0kJyI8z5|H`Q?~^A!|S-=LT)Gk%rd(Dn2cJB7@`Wd4a8)>t2U8e znkT$zD65>ekyJ$m%)q287D_NAl$#jQ87nl ziQ8C4&eB6Tc?Jvl9+?KUe8<_T=9vbfr9)Ne0C(O?vp0v#NkCzenS73GmnxP`2?5^} zPa@$(FboSSp&;EJ_@Sa5UJbu79K9Zbx6<-e;7IQG&Q1v{I54V1`s|kdUoIaNv*A6` z^OKMe0)WUuRx=xke#VekVlDSNmPt;rt`3}B#EhcbuEcm+#ZA9kHw$q|plShB*z!z~ zig4Tc*tP0oVobYQ3Z1Yl@loO<)5^8M!8fs56q>R41LWBB=g}B_VkwXirn`Oz0`)8- zXKX8)*b`0b^f%`gtTe0rL|OLalb3O>Ug_nsGIPhfqvAI$nET-yyROc=#2?|5Z13Sw zGZ^{mk&F7^v{}dcI6rh4e(FJ5PTR>TOD26QzvxsnEiv8@Qczg+K4&pRve~~*eD{VA zv9blm0pPHntKX|v*2z;|QQW<#FTdrckH2xu@z~&8{U_z67iv7zXHSnfS+AH9;k<9x zajJijI5KC71V~;o`18P|kBa#*{s|@>icAxKDVq1TJS`{vdvQf20Cj#@K)Qg4VeI8e ztI-9E7IY37p!?3^>?6mIA8&J=gQc5k0MN3l;?(>nRaGz7n|KL4_qY10qK*B6gFB~D zaFFrKKY4Y|)q51qr?R>Hu?cnZS7VM0e_zJYy{^A{w`xHts0tsOg!|CCamGPBUPNO&kVV8>=w+AVivCU5;za zsF;Fb!-rQx;WLFX2*)sbW{mhtg*9rfJm@jTHvM-qP7r=6#gi~~f7Kp)K24Vs#Tmvo zMEExO?#_@R!}fu>>s+e9d|Aw{q1JD%kYs1DY3Q;O43}2<;c6wiHcESIXt!H6Wv`Fa zF`E0$PUfM!;9^*L7Q@)^!4p7`WXQKl=2kMX41o~`r=Siypr_(@Wcjo8j^4sXD8dIQ zv?{tq7GEvm!LS)uHmMhJ9({0kKoMXeFi+)2(xc&xSk0k)R?6F!qHB=9U|ymRp;wrg zz=F2)KQ(g}?lhdePCz|kf@N4rif`PpBc+V##LfoHz4+4)#AdH^!mJY0Zy7Z(k&DBN z;x0x{G=kkM{h)rkSG~A%VklHe8@|+p6ptyE`G^!{a$InsFIdBGY|~w z_vzUmLXQEzb)#WvZ|-+8`3wc2sEcc=2EHp@DAi~;0>Sy1!(g-Zj6dqKjcz=pI?ZP? zr)epGBky;cNmNw5$Q>mx5V-8N0IhYJ4^Ollq6(u?J-g#y$1dNgr@WcHbHv%(z<~+- z_HD@M_!vbJ*KXrB)rQ=EHyU12+E(FO5cBDK4-u#6=CPvV5fs5ca;>RpH4(JGeA~nh zy_yi@u7}&V;W;jlbMxX)^RhH#jJ81~0a$pbHfhHE2bF+}=Q~ zv#j$RNeu3vy^q?I*S`<)k#RGx{~)&WUslh2MXnb3`aR7&=X9c<*}7*ZoV@YD+l413Cs7-i& zU7M-fP^bw1f!^h(#*O3ATNStJPF1ivvx^G*+oY31)Fg=?FikwCcH)u&>dhP-H0-Gy z0`Os5W{8;K4kl_%fP_KoX08%+-({Q74_V=J7=T|2(5Y|MGlq~(oPJ0c znv8<0aeEj`TjGuS-Iah@e1@K&laXr2te0%#5kNhHHio%V51J#f=79y73_b%9DAl{f z4z3YLPzMpRTpwmD*ldRiM#ktu2O@>|jeUay#YNlv<@q1Qqy{%=(Z^6_jiF2(2mTLDrs&rp<31vj$ju};a=o} z=E<~HICJ8y216ds#4GO-S7NI~3ra#4kP{iw%6$+{Oc(%BYV%F&cw|mzN?*<7ON^_< zE*F!CI67ew&ZS}T1@V82M+a<8{3u9##9U>~nz__}vgc2!WB)^ttE+LyK`UsX^bGFg zRmC-7%_aL&0e!8BsfP+*6QQ2UWErvsC3>WX84|aNuw?{DmB3(Bl{59o#6xrew8i%39lF6>AcFk2_JfajI*Cf7`uccYdteE8y^-<*5-2FOS39h2sEvx@~@s|eFmZZ8q z#%!_SwMWCpFCKPj?em*&bOsFg({F0or-$XaOMhB^_+9%XPmc=MVowiHil{#i)Si3s zCA?lSKPT`42~`wV93AJhzW}=By@gE!i7r23*WQ4Jl_430(KUs0V*;C6U-603vq|$X zDR_K+R6s*+1fb6rO{obt*euv?Hmk-Xw3#isDp{q~TUY^ZYhl(|F@GUi}Yh0IxY+_8+nI zDo%K#>r}v;@gGL)qQ#+bh>WcRBC*k(&&-$2QFWV}O`5dk-skOe#n_A?JSHquJ@0?g zyt>k??h|0$B&l=v?!~IU^ZJV1Pst{I2mX2K^wyM$#|x_$xz@Gz(7WM2Py@6_>Ur9h zTB+1XpehdfTHlM?l7u4DzuIiuSZ(aT&%11nQ@E+2R^PhTs%b`AU%Ui@C9ssuu}Aq$ zJ0YO0v|nBo*SbR^1PEWGIZTI5e9xKjp8QcKtN!;Igu?SOT{eeM?5EW&g(2!&2S!4YP?c(w;Zt;k}Zl7!@rD7yCDy6Aqm;g+One47VeDI(a z)}>|rJ;!eOZ^SH6iVxf*;gu}By@1{@6jRTh;=m<_U65l6=M)3T9-&kbbvZN+%ajDd zLL9^vsY}_`P+Ks5%gXEG3T~i#zh$P0 zUqR?bqFSQxqfCT~ElQfQM~-4EBUHMnX0;HhvkaPJ5oL*cucfD%+zG&SwH9BAJp@Bh zdKXb7*Fs7z(e)Qv4JwGJM5IrZh!=0sRDTU?eVwm==GJvBQ*H1HMoh^(Aqg*9OAsqp z1);Nw&^^pJ$%V1Fn7C3qOT8IOT zr@132YEwnUvGlC6Wh6a~Im$#JcCoD2dsAC$owDh?w!jrF$N%|pJeq6z_=PHQX3n{) z>4{&R+y;C+q7qB-H~9bnpo&$){z4ZNr-f@EAlcgV+^<_rZA8(@>wiQ!`A?|S{vS(y zQ%xjiDRj~?4Qb*d*kV`suWCOm@4w-HLgiYT?Iv0oR>F#w4pvhq29pO*5l*}|A!Cji z5a0mbi#F8DkFVLuf16?04CYnhtO==L0^PdETUmCpd4}?6S4quEOcuvu_y*xts=uvl zOpw_b620hrB?^tHN}HUNp@xVbzoo4Bg=r@X3xu#|*nFrYw~(*De+_iwcu+MVDG2L@ z!G%x_=$vIfD%CFVPlcU)tdAzW!|s!Rq}U;s?2&;aUrF{B3)PdvG)7FysDxu?)yRC} zAxE)DOgquQvggJ|jyD-XTpYmCo^rNt5WC?+fV3zEh^?i5< zTkBe8=1{63tCWo!OnpDEbe&gWyye{IAwV(oxLYsHWc`5ee$wxb*E#(l+(6w_Tf2cj zEIyF;;w;nw+i1rnj*FrliOU5b8>8m;{orr5|Ha!!uc_jnqWk8i4LOyvWeY`2?zbYQ z4Ev47U5@rM+h=Cu@wwym(UVh(AHJuHmqy;c=?{+tmo>a6axil9%gtjI^@xFAFW-}AP8)}_@P2K+VN_4*MO9f zDqjNo#>W&m4F^g}0!m7+b2w&`eIA&>87B$u0Y`Rv&82bEl^oq5qm-f6x(|6PEAX|&p{!{atIsK z%`n2T5j{B&wtIxE8j`#JJmcL|S>Q4JE(L>0W~|9qvXKn5c)nzVhvZPGUu-TYy;9gEQekE9j* zr>}bP@y{>O#ZN0lOsN|1pje9=X7_hwbm0JqATiv3@nUTajXylD&YDv)`0b3DZJo5f zUvF0?|Mjq1Yp!8`mcZ7#JK+I{lm+?eJttdv?hhI|HRZXN|Dk?>)f6U8naDu?mLtWM z4HA1|dhezPkI_``cTl*)#xfP!5WP4I|cFJXyI)PeQJYXr$n~5n@Nx@9n1E)Y`SQGn!oI zHC)f-sxPtq6H$_g2x8Q(ORpSY*Y;Rf|x1Dk~@AHp!rxwb3b9}6sPUFng*^ON2nBq z2(Y8*qvjPSb{$z4wrlS_r2mxjdVEaDG4s#73ZiyPY;#Xm39^R?%>a#k&yn*=;qFKk zH;xNY2gly5w4>@m{9HwY1`r^Y9(;j=aPtn@_1(XuwKU%{x`gM}11g>jf;q(_!-U_) zy(TLOB`@jECmsR)gHE?qc(N9AXqlcM#G+>#M2~!`T3;6ml3$I3Zd+2pVI03eCU%U< zU^#{$+Tw7)2p%HmY#G9U#HJkfWeb{caYP1BDmw_vrHbh|^}qY67y6-5lQBnN6h#`` zFZK#@7*V+1;-eW&d4Ztp0AnzEUWxuB&D%7?No}TKH7={Ik!HBjs3_IUO%%@qmida$ z7(#0&tEdqjEQ8F%U{1)sU)`cu;R=QM2pwk?l`Eg@;pG-8K^mC`ZGQcAzsMWJx-!Qo zTSx)2(-IFxTZV9=7wrS^E8E^2t{60zf!MC4uh$9}QPj79szOo0uS6vxx|vu|%4p!r zu?+u~>*79rQTROdd_^~_-6xcCw^<%gY_Kw7meq)0-&=!QZ5h{i?GHx3b!c%?XK!ps z$du!nmj+!|xw4|+CF|ao9QWF|t!t#{+O$WztqPWRZoFved}Gs(ip>%tk7Z0V`q=nJ zV6MrHB!8tR#WzfzENP{xDjsR~q1WGZOI9{^S4N3*_jx}B)^Hdzk~vPbK(g+nq0sug{9Cs+ zM@6j%y6-%2pgKb=>bd?1@=K1^+fJCkUIqnwd;2@*mJLL)we_91Y0S94&K+1a{9cBC zJE}KjLnlR=IC(*G@e)=~|s%MrwNc`uO;{QPuBZDG%G5H^SC- zYjChNV+Mf2(-R|$&jzF`gig1)-E&-n;ua^7agH;wJ#IsZQHkJ!RT@<8_kXFYG{~v>D`W_N#?VM74+f>Ypw7z zn!_IFU_0^;EjjB^)ynY3&p^?>c8^FBK^LBKdp4-bKAu|}2Zy*GH?p(aD1>d5YBT?X3T#Ka4P zDOp5aP*9+(=3M^Z!AeT{_?grEYx*dhr=_@iq=<5+7nVZLN>Benipr!_j=F@Uj@9YJ z+l3lM^K|v3g zA$MC69TePg=YcDMNcisK?`%4Wzl1v2p2B~1LPCPqRxcWEZEfwb56D`O+Zru}@+eEH z9ZRQLSTsV@r=+E&W&0txb(h_FE_cSne%bKsnY*2%HyR~joRDQ!c`?PeX6pKnvj`CPPU5p&<=s4ud!v%_4Nd5lCN^esx z!i0$v+p9U>yc<|lTYM*ijG&>V)fl0xGGbcKjlG5t^(g1%ifn^W@19ZM859to9B`ly z+|Eoe@6C~sjhP=q0{a!X`XV`F)YM6x4KqHfecX8CSIDn+CtX@I`j+rmU+)*J2nHWM zps9j;S=z6(nS&=!Ru#r=?enq@V8D|KZ9y2hW-}>3t<*2w9K$=T2tK}R$QKog0KJaO z+HbV*Y2Lnr^YrZbn+l2NyAu4i)TCZ{QdR>MCT3%7{JJ@`bdyJE6HaDf$GI2BjZki|il_N>4~aZjE=61aH2v0UM=pk8GlF6V}F^ zyLa~-J*re!SLfp5vYfzXxK8!ct5*-o%N5E$e_XqLyNYDxy${<*Cbwz)=}TZyd#V`u zhLx3-#9`N0Ro=rdI}v(^1_YQ(q3{A5dpRRFU3%MzHBnucTfLsOOsFxM2Z!N2!7v11 zPoqztUz3xQm5t^!_ww?xcKEl;3SVwpZ`Ys64=rLwAJkkM^>#dZ>$L?16If&io2(s! zgv;EVt!9;c%RE3Q5}tNI_<_H|son;rUjEtKLRD4ux&O`#yc4|o#-8wO^yO5`RJdiF zn9lwBDMRUS?RB8bEW_}vTP3^l5RLt_Z)ptZR0CoKw<^VcpxZ#XzT3BN&$&kX-T-%y z+qSMlE$$6ER&?*(`9{mZQ)e(dz+be5o15DL@Nf8eUT z{OF!N7gi^9s=xQ}#fwd(q1ndWRzZV|J#}iLNu!M$ja>Y!tLywYiZfvqA5Te{pEQ*` zbAfJ+%eI19^HN$KHJ2$1dCYR^mj&7D6Nx)@ehMGl^3-o`)`tC8ATZhpHv}!_PxRKm zd3*P!&Ra!MsMCu8H^|oZS2|12>Owu}wlhhG3m_dN$oe38Y*}!&2|iFcwIT&eP(p$p z{bNG~#imU!a0^>^>=;6A5PR}u8(`JGf;C^fe7UoR#{Jk+yBvz>ssMGB8ONzjJYu&Z zG0{*C1OZ-KUw_hBeaZy5KCOQFWmReEV6M7NS4VpWFX!# z>*>{P&1>JQn0jkdYOG<5)7jiy&DPeopI@yugP2}yL@Y-yRb%3Y~-^7t7uG;C~a7Vy=hO?{d&q+H_k2jcHU1CKLfgP5|{!5inYj+Mqqq7pN7_u3$eVSlvt6^pao8wg@@`M4) z9~WIx>1Sp(i&>`$0vzfSEJDZ4viF$R}-6MNve7H=@ z*3~p-G~-uWqu;hQb&gzok^}8T37pSkLx|oC->18-uDj0C zVNWo}M6syzLfWtyxpXjbO<`=nz33?`pWZjh?#b?VT`p#8&=h$$(y3Ko;MpFh;^M5x z9m5XZT1%3qUd^I2?S(#s;+3`+PVGBF5=eK z1seGgqSUL}+Hg{+arK?WA1*W5=}ID^!I)t5c>51rkAA&7cizXOKdG4|Mi}8S-gZT@2b^Al}M*@Sqwm00;{M#Q6bNX3z%ZlJ{ zFYA@Qh7bkiwb{}jYtZ+9@_+k5>;H~_{7L;#?Lj~9^1^)rlUx2adf3q+(^vir3)jVt diff --git a/results-loop/lcb_labels.json b/results-loop/lcb_labels.json new file mode 100644 index 0000000..f092297 --- /dev/null +++ b/results-loop/lcb_labels.json @@ -0,0 +1,152 @@ +{ + "abc382_a": true, + "abc382_b": true, + "abc382_c": false, + "abc382_d": true, + "abc382_f": false, + "abc382_g": false, + "abc383_a": true, + "abc383_b": true, + "abc383_c": false, + "abc383_d": true, + "abc383_e": false, + "abc384_a": true, + "abc384_b": true, + "abc384_c": true, + "abc384_d": false, + "abc384_e": false, + "abc384_f": false, + "abc384_g": false, + "abc385_a": true, + "abc385_b": true, + "abc385_c": false, + "abc385_d": false, + "abc385_e": false, + "abc385_f": false, + "abc386_a": false, + "abc386_b": true, + "abc386_c": false, + "abc386_d": false, + "abc386_e": false, + "abc386_f": false, + "abc387_a": true, + "abc387_b": true, + "abc387_c": false, + "abc387_f": false, + "abc388_a": true, + "abc388_b": true, + "abc388_c": false, + "abc388_d": false, + "abc388_e": false, + "abc388_f": false, + "abc388_g": false, + "abc389_a": true, + "abc389_b": true, + "abc389_d": false, + "abc389_e": false, + "abc389_f": false, + "abc389_g": false, + "abc390_a": true, + "abc390_b": true, + "abc390_c": false, + "abc390_d": false, + "abc390_e": false, + "abc390_f": false, + "abc390_g": false, + "abc391_a": true, + "abc391_b": true, + "abc391_d": false, + "abc391_e": false, + "abc391_f": false, + "abc391_g": false, + "abc392_a": true, + "abc392_b": true, + "abc392_c": false, + "abc392_d": false, + "abc392_f": true, + "abc392_g": false, + "abc393_a": true, + "abc393_b": true, + "abc393_d": false, + "abc393_e": true, + "abc393_f": false, + "abc394_a": true, + "abc394_b": true, + "abc394_c": true, + "abc394_d": true, + "abc394_e": false, + "abc394_f": false, + "abc394_g": false, + "abc395_a": true, + "abc395_b": true, + "abc395_c": true, + "abc395_e": true, + "abc395_f": false, + "abc396_a": true, + "abc396_b": true, + "abc396_c": false, + "abc396_d": true, + "abc396_e": false, + "abc396_f": false, + "abc396_g": false, + "abc397_a": true, + "abc397_b": false, + "abc397_c": true, + "abc397_d": false, + "abc397_e": false, + "abc397_f": false, + "abc397_g": false, + "abc398_a": true, + "abc398_b": false, + "abc398_c": true, + "abc398_d": false, + "abc398_f": true, + "abc398_g": false, + "abc399_a": true, + "abc399_b": true, + "abc399_c": true, + "abc399_d": false, + "abc399_e": false, + "abc399_f": false, + "abc400_a": true, + "abc400_b": true, + "abc400_c": false, + "abc400_d": false, + "abc400_e": false, + "abc400_g": false, + "arc188_a": false, + "arc188_b": false, + "arc188_c": false, + "arc188_d": false, + "arc189_a": false, + "arc189_b": false, + "arc189_c": false, + "arc189_d": false, + "arc190_a": false, + "arc190_c": false, + "arc190_d": false, + "arc191_a": false, + "arc191_c": false, + "arc191_d": true, + "arc192_a": false, + "arc192_b": false, + "arc192_d": false, + "arc192_e": false, + "arc193_a": false, + "arc193_b": false, + "arc193_d": false, + "arc194_a": false, + "arc194_b": false, + "arc194_c": false, + "arc194_d": false, + "arc194_e": false, + "arc195_a": false, + "arc195_b": false, + "arc195_c": false, + "arc195_d": false, + "arc195_e": false, + "arc196_a": false, + "arc196_b": false, + "arc196_c": false, + "arc196_d": false +} \ No newline at end of file diff --git a/scripts/figures_final.py b/scripts/figures_final.py index bb48d31..69b4a13 100644 --- a/scripts/figures_final.py +++ b/scripts/figures_final.py @@ -220,8 +220,14 @@ def fig_transfer(): ("base", *ev(OUT / "eval_rust_py_transfer.json", 0)), ("loop k=4", *ev(OUT / "eval_rust_py_transfer.json", 4)), ]), + ("LiveCodeBench (far)\nuntrained wins", [ + ("base", *he_hard("eval_lcb_untrained.json", 0)), + ("untrained k=4", *he_hard("eval_lcb_untrained.json", 4)), + ("loop k=4", *he_hard("eval_lcb_loop.json", 4)), + ("distill", *he_hard("eval_lcb_distill.json", 1)), + ]), ] - fig, axes = plt.subplots(1, 2, figsize=(7.6, 3.8)) + fig, axes = plt.subplots(1, 3, figsize=(10.8, 3.8)) for ax, (title, arms) in zip(axes, groups): style(ax) x = range(len(arms))