run_id string | iteration int64 | cumulative_tokens int64 | best_val_score float64 | accepted bool | candidate_idx int64 | candidate_prompt string | best_prompt string |
|---|---|---|---|---|---|---|---|
fixed_iter_rlm_k20 | 1 | 134,217 | 0.466667 | true | 1 | Answer construction and checking checklist:
A. Final answer format
- Print only the final answer on the last line as: ### <answer> (no extra text) [Ex 1–14].
- Preserve required leading zeros to 3 digits when that convention is used (e.g., ### 021, ### 073, ### 033) [Ex 5, Ex 11, Ex 14].
- When the prompt asks for m+n... | Answer construction and checking checklist:
A. Final answer format
- Print only the final answer on the last line as: ### <answer> (no extra text) [Ex 1–14].
- Preserve required leading zeros to 3 digits when that convention is used (e.g., ### 021, ### 073, ### 033) [Ex 5, Ex 11, Ex 14].
- When the prompt asks for m+n... |
fixed_iter_rlm_k20 | 2 | 405,119 | 0.466667 | true | 2 | New Instructions for the Assistant
1) Final output format (Examples 1, 7): Produce exactly one line containing the answer in the form '### <answer>', with no other text, explanations, headings, or lines before or after.
2) Three-digit padding (Examples 3, 8, 10, 13, 9): When the answer is an integer in AIME-style 0–9... | Answer construction and checking checklist:
A. Final answer format
- Print only the final answer on the last line as: ### <answer> (no extra text) [Ex 1–14].
- Preserve required leading zeros to 3 digits when that convention is used (e.g., ### 021, ### 073, ### 033) [Ex 5, Ex 11, Ex 14].
- When the prompt asks for m+n... |
fixed_iter_rlm_k20 | 3 | 687,283 | 0.466667 | false | 2 | New Instructions for the Assistant
1) Final output format (Examples 1, 7): Produce exactly one line containing the answer in the form '### <answer>', with no other text, explanations, headings, or lines before or after.
2) Three-digit padding (Examples 3, 8, 10, 13, 9): When the answer is an integer in AIME-style 0–9... | Answer construction and checking checklist:
A. Final answer format
- Print only the final answer on the last line as: ### <answer> (no extra text) [Ex 1–14].
- Preserve required leading zeros to 3 digits when that convention is used (e.g., ### 021, ### 073, ### 033) [Ex 5, Ex 11, Ex 14].
- When the prompt asks for m+n... |
fixed_iter_rlm_k20 | 4 | 894,987 | 0.466667 | false | 2 | New Instructions for the Assistant
1) Final output format (Examples 1, 7): Produce exactly one line containing the answer in the form '### <answer>', with no other text, explanations, headings, or lines before or after.
2) Three-digit padding (Examples 3, 8, 10, 13, 9): When the answer is an integer in AIME-style 0–9... | Answer construction and checking checklist:
A. Final answer format
- Print only the final answer on the last line as: ### <answer> (no extra text) [Ex 1–14].
- Preserve required leading zeros to 3 digits when that convention is used (e.g., ### 021, ### 073, ### 033) [Ex 5, Ex 11, Ex 14].
- When the prompt asks for m+n... |
fixed_iter_rlm_k20 | 5 | 1,023,988 | 0.466667 | false | 2 | New Instructions for the Assistant
1) Final output format (Examples 1, 7): Produce exactly one line containing the answer in the form '### <answer>', with no other text, explanations, headings, or lines before or after.
2) Three-digit padding (Examples 3, 8, 10, 13, 9): When the answer is an integer in AIME-style 0–9... | Answer construction and checking checklist:
A. Final answer format
- Print only the final answer on the last line as: ### <answer> (no extra text) [Ex 1–14].
- Preserve required leading zeros to 3 digits when that convention is used (e.g., ### 021, ### 073, ### 033) [Ex 5, Ex 11, Ex 14].
- When the prompt asks for m+n... |
fixed_iter_rlm_k20 | 6 | 1,150,570 | 0.466667 | false | 2 | New Instructions for the Assistant
1) Final output format (Examples 1, 7): Produce exactly one line containing the answer in the form '### <answer>', with no other text, explanations, headings, or lines before or after.
2) Three-digit padding (Examples 3, 8, 10, 13, 9): When the answer is an integer in AIME-style 0–9... | Answer construction and checking checklist:
A. Final answer format
- Print only the final answer on the last line as: ### <answer> (no extra text) [Ex 1–14].
- Preserve required leading zeros to 3 digits when that convention is used (e.g., ### 021, ### 073, ### 033) [Ex 5, Ex 11, Ex 14].
- When the prompt asks for m+n... |
gepa-rlm-exp-failures_only-20260219-191545
GEPA prompt optimization experiment on AIME math problems.
Task LM: openai/gpt-4.1-mini | Reflection LM: openai/gpt-5 | Reflection Mode: failures_only | Last updated: 2026-02-19 21:50 UTC
Results
| Run | Method | k | Mode | Val Score | Test Acc | Tokens | Cost | Time |
|---|---|---|---|---|---|---|---|---|
| fixed_rlm_k20 | rlm | 20 | failures_only | 46.67% | 30.67% | 1,150,570 | $0.0000 | 5996s |
Learning Curves
Experiment Config
{
"script_name": "run_experiment.py",
"model": "openai/gpt-4.1-mini",
"reflection_lm": "openai/gpt-5",
"reflection_mode": "failures_only",
"hyperparameters": {
"task_lm": "openai/gpt-4.1-mini",
"reflection_lm": "openai/gpt-5",
"reflection_mode": "failures_only",
"seed_prompt": "You are a helpful assistant. You are given a question and you need to answer it. The answer should b..."
},
"input_datasets": [
"AI-MO/aimo-validation-aime",
"MathArena/aime_2025"
],
"description": "GEPA prompt optimization experiment on AIME (reflection_mode=failures_only)",
"num_runs": 1,
"last_updated": "2026-02-19 21:50 UTC"
}
Dataset Configs
This repo contains multiple configs (subsets). Load them with:
from datasets import load_dataset
results = load_dataset("reasoning-degeneration-dev/gepa-rlm-exp-failures_only-20260219-191545", "results", split="train")
traces = load_dataset("reasoning-degeneration-dev/gepa-rlm-exp-failures_only-20260219-191545", "traces", split="train")
val_traces = load_dataset("reasoning-degeneration-dev/gepa-rlm-exp-failures_only-20260219-191545", "val_traces", split="train")
curves = load_dataset("reasoning-degeneration-dev/gepa-rlm-exp-failures_only-20260219-191545", "curves", split="train")
| Config | Description |
|---|---|
results |
Per-run summary: method, k, val/test scores, tokens, cost, optimized prompt |
traces |
Per-example test set traces with model responses and scores |
val_traces |
Per-iteration validation traces with the prompt used at each step |
curves |
Per-iteration learning curves: tokens, scores, candidate prompts |
State file: state.json
For experiment design details, see EXPERIMENTS-DESIGN.md
- Downloads last month
- 32