| 123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142 |
- {
- "cot_reasoning": {
- "timestamp": "2026-03-14T22:32:25.567957",
- "ability_name": "cot_reasoning",
- "r_eff_curve": [
- 3.7347538471221924,
- 4.651706218719482,
- 3.4198639392852783,
- 3.5962984561920166,
- 6.821524143218994
- ],
- "model_order": [
- "llama3.2-3b-instruct",
- "Mistral-7B-v0.3",
- "llama3-8b",
- "gemma-2-9b-it",
- "Qwen2.5-7B-Instruct"
- ],
- "model_mmlu": [
- 58.0,
- 62.5,
- 68.4,
- 82.0,
- 86.3
- ],
- "max_jump": {
- "model_index": 3,
- "model_name": "gemma-2-9b-it",
- "mmlu_at_jump": 82.0,
- "jump_magnitude": 3.2252256870269775
- },
- "changepoints": [
- 5
- ],
- "jump_significance": 2.5551615281279734
- },
- "arithmetic": {
- "timestamp": "2026-03-14T22:35:34.370517",
- "ability_name": "arithmetic",
- "r_eff_curve": [
- 4.50555419921875,
- 5.898599624633789,
- 5.256784439086914,
- 4.236079216003418,
- 13.66690731048584
- ],
- "model_order": [
- "llama3.2-3b-instruct",
- "Mistral-7B-v0.3",
- "llama3-8b",
- "gemma-2-9b-it",
- "Qwen2.5-7B-Instruct"
- ],
- "model_mmlu": [
- 58.0,
- 62.5,
- 68.4,
- 82.0,
- 86.3
- ],
- "max_jump": {
- "model_index": 3,
- "model_name": "gemma-2-9b-it",
- "mmlu_at_jump": 82.0,
- "jump_magnitude": 9.430828094482422
- },
- "changepoints": [
- 5
- ],
- "jump_significance": 2.6749716525969935
- },
- "code_generation": {
- "timestamp": "2026-03-14T22:40:01.844139",
- "ability_name": "code_generation",
- "r_eff_curve": [
- 6.77747917175293,
- 14.093628883361816,
- 9.412063598632812,
- 7.725470542907715,
- 4.795677661895752
- ],
- "model_order": [
- "llama3.2-3b-instruct",
- "Mistral-7B-v0.3",
- "llama3-8b",
- "gemma-2-9b-it",
- "Qwen2.5-7B-Instruct"
- ],
- "model_mmlu": [
- 58.0,
- 62.5,
- 68.4,
- 82.0,
- 86.3
- ],
- "max_jump": {
- "model_index": 0,
- "model_name": "llama3.2-3b-instruct",
- "mmlu_at_jump": 58.0,
- "jump_magnitude": 7.316149711608887
- },
- "changepoints": [
- 5
- ],
- "jump_significance": 2.3277893513824783
- },
- "instruction_following": {
- "timestamp": "2026-03-14T22:41:56.391270",
- "ability_name": "instruction_following",
- "r_eff_curve": [
- 7.395653247833252,
- 16.79442596435547,
- 15.557455062866211,
- 11.049053192138672,
- 12.142755508422852
- ],
- "model_order": [
- "llama3.2-3b-instruct",
- "Mistral-7B-v0.3",
- "llama3-8b",
- "gemma-2-9b-it",
- "Qwen2.5-7B-Instruct"
- ],
- "model_mmlu": [
- 58.0,
- 62.5,
- 68.4,
- 82.0,
- 86.3
- ],
- "max_jump": {
- "model_index": 0,
- "model_name": "llama3.2-3b-instruct",
- "mmlu_at_jump": 58.0,
- "jump_magnitude": 9.398772716522217
- },
- "changepoints": [
- 5
- ],
- "jump_significance": 2.8074494163513304
- }
- }
|