eval-Hermes-4-14B-nonreasoning

# h4-14b-more-stage1-nonreasoning Evaluation Results

## Summary

| Benchmark | Score | Metric | Samples | Overlong rate |
|-----------|-------|--------|---------|---------------|
| aime24 | 0.110 | math_pass@1:64_samples | 64 | 99.9% |
| aime25 | 0.069 | math_pass@1:64_samples | 64 | 99.6% |
| arenahard | 0.502 | eval/overall_winrate | 500 | 0.0% |
| bbh_generative | 0.740 | extractive_match | 1 | 100.0% |
| creative-writing-v3 | 0.355 | creative_writing_score | 96 | 0.0% |
| drop_generative_nous | 0.739 | drop_acc | 1 | 100.0% |
| eqbench3 | 0.580 | eqbench_score | 135 | 0.0% |
| gpqa_diamond | 0.390 | gpqa_pass@1:8_samples | 8 | 99.7% |
| ifeval | 0.788 | inst_level_loose_acc | 1 | 99.8% |
| lcb-v6-aug2024+ | 0.350 | eval/pass_1 | 1 | 50.7% |
| math_500 | 0.763 | math_pass@1:4_samples | 4 | 100.0% |
| mmlu_generative | 0.792 | extractive_match | 1 | 100.0% |
| mmlu_pro | 0.619 | pass@1:1_samples | 1 | 100.0% |
| musr_generative | 0.525 | extractive_match | 1 | 99.9% |
| obqa_generative | 0.904 | extractive_match | 1 | 100.0% |
| rewardbench | 0.540 | eval/percent_correct | 1 | 94.1% |
| simpleqa_nous | 0.043 | fuzzy_match | 1 | 100.0% |

Overlong rate: 60,082 / 64,523 samples (93.1%) missing closing `</think>` tag

## Detailed Results

### aime24

| Metric | Score | Std Error |
|--------|-------|----------|
| math_pass@1:1_samples | 0.067 | 0.046 |
| math_pass@1:4_samples | 0.117 | 0.053 |
| math_pass@1:8_samples | 0.092 | 0.043 |
| math_pass@1:16_samples | 0.104 | 0.044 |
| math_pass@1:32_samples | 0.106 | 0.045 |
| math_pass@1:64_samples | 0.110 | 0.044 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:07:48
**Temperature:** 0.6
**Overlong samples:** 99.9% (1918 / 1920)

### aime25

| Metric | Score | Std Error |
|--------|-------|----------|
| math_pass@1:1_samples | 0.033 | 0.033 |
| math_pass@1:4_samples | 0.033 | 0.020 |
| math_pass@1:8_samples | 0.054 | 0.026 |
| math_pass@1:16_samples | 0.065 | 0.025 |
| math_pass@1:32_samples | 0.074 | 0.029 |
| math_pass@1:64_samples | 0.069 | 0.027 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:07:15
**Temperature:** 0.6
**Overlong samples:** 99.6% (1912 / 1920)

### arenahard

| Metric | Score | Std Error |
|--------|-------|----------|
| eval/overall_winrate | 0.502 | 0.000 |
| eval/total_samples | 500.000 | 0.000 |
| eval/win_count | 199.000 | 0.000 |
| eval/tie_count | 102.000 | 0.000 |
| eval/loss_count | 199.000 | 0.000 |
| eval/win_rate | 0.398 | 0.000 |
| eval/tie_rate | 0.204 | 0.000 |
| eval/loss_rate | 0.398 | 0.000 |
| eval/winrate_arena-hard-v0.1 | 0.502 | 0.000 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:02:37
**Temperature:** 0.6
**Overlong samples:** 0.0% (0 / 500)

### bbh_generative

| Metric | Score | Std Error |
|--------|-------|----------|
| extractive_match | 0.740 | 0.026 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:22:26
**Temperature:** 0.6
**Overlong samples:** 100.0% (5511 / 5511)

### creative-writing-v3

| Metric | Score | Std Error |
|--------|-------|----------|
| creative_writing_score | 0.355 | 0.000 |
| num_samples | 96.000 | 0.000 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** N/A
**Temperature:** N/A
**Overlong samples:** 0.0% (0 / 96)

### drop_generative_nous

| Metric | Score | Std Error |
|--------|-------|----------|
| drop_acc | 0.739 | 0.004 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:37:37
**Temperature:** 0.6
**Overlong samples:** 100.0% (9536 / 9536)

### eqbench3

| Metric | Score | Std Error |
|--------|-------|----------|
| eqbench_score | 0.580 | 0.000 |
| num_samples | 135.000 | 0.000 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** N/A
**Temperature:** N/A
**Overlong samples:** 0.0% (0 / 135)

### gpqa_diamond

| Metric | Score | Std Error |
|--------|-------|----------|
| gpqa_pass@1:1_samples | 0.409 | 0.035 |
| gpqa_pass@1:4_samples | 0.383 | 0.025 |
| gpqa_pass@1:8_samples | 0.390 | 0.023 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:07:08
**Temperature:** 0.6
**Overlong samples:** 99.7% (1579 / 1584)

### ifeval

| Metric | Score | Std Error |
|--------|-------|----------|
| prompt_level_strict_acc | 0.638 | 0.021 |
| inst_level_strict_acc | 0.728 | 0.000 |
| prompt_level_loose_acc | 0.713 | 0.019 |
| inst_level_loose_acc | 0.788 | 0.000 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:07:41
**Temperature:** 0.6
**Overlong samples:** 99.8% (540 / 541)

### lcb-v6-aug2024+

| Metric | Score | Std Error |
|--------|-------|----------|
| eval/pass_1 | 0.350 | 0.000 |
| eval/easy_pass_1 | 0.863 | 0.000 |
| eval/medium_pass_1 | 0.358 | 0.000 |
| eval/hard_pass_1 | 0.068 | 0.000 |
| eval/completion_length | 23308.832 | 0.000 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 01:04:40
**Temperature:** N/A
**Overlong samples:** 50.7% (3681 / 7264)

### math_500

| Metric | Score | Std Error |
|--------|-------|----------|
| math_pass@1:1_samples | 0.764 | 0.019 |
| math_pass@1:4_samples | 0.763 | 0.016 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:02:27
**Temperature:** 0.6
**Overlong samples:** 100.0% (2000 / 2000)

### mmlu_generative

| Metric | Score | Std Error |
|--------|-------|----------|
| extractive_match | 0.792 | 0.003 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:55:09
**Temperature:** 0.6
**Overlong samples:** 100.0% (14042 / 14042)

### mmlu_pro

| Metric | Score | Std Error |
|--------|-------|----------|
| pass@1:1_samples | 0.619 | 0.004 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:47:17
**Temperature:** 0.6
**Overlong samples:** 100.0% (12032 / 12032)

### musr_generative

| Metric | Score | Std Error |
|--------|-------|----------|
| extractive_match | 0.525 | 0.031 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:02:59
**Temperature:** 0.6
**Overlong samples:** 99.9% (755 / 756)

### obqa_generative

| Metric | Score | Std Error |
|--------|-------|----------|
| extractive_match | 0.904 | 0.013 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:01:59
**Temperature:** 0.6
**Overlong samples:** 100.0% (500 / 500)

### rewardbench

| Metric | Score | Std Error |
|--------|-------|----------|
| eval/percent_correct | 0.540 | 0.000 |
| eval/total_samples | 1865.000 | 0.000 |
| eval/correct_samples | 1008.000 | 0.000 |
| eval/format_compliance_rate | 0.948 | 0.000 |
| eval/avg_response_length | 2973.564 | 0.000 |
| eval/response_length_std | 6132.534 | 0.000 |
| eval/judgment_entropy | 1.465 | 0.000 |
| eval/most_common_judgment_freq | 0.386 | 0.000 |
| eval/format_error_rate | 0.055 | 0.000 |
| eval/avg_ties_rating | 4.218 | 0.000 |
| eval/ties_error_rate | 0.365 | 0.000 |
| eval/percent_correct_Factuality | 0.402 | 0.000 |
| eval/percent_correct_Precise IF | 0.244 | 0.000 |
| eval/percent_correct_Math | 0.519 | 0.000 |
| eval/percent_correct_Safety | 0.578 | 0.000 |
| eval/percent_correct_Focus | 0.711 | 0.000 |
| eval/percent_correct_Ties | 0.696 | 0.000 |
| eval/choice_samples | 1763.000 | 0.000 |
| eval/ties_samples | 102.000 | 0.000 |
| eval/choice_format_compliance_rate | 0.945 | 0.000 |
| eval/ties_format_compliance_rate | 1.000 | 0.000 |
| eval/wrong_answer_a_bias_rate | 0.472 | 0.000 |
| eval/wrong_answer_total_count | 826.000 | 0.000 |
| eval/wrong_answer_a_count | 390.000 | 0.000 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:03:06
**Temperature:** 0.6
**Overlong samples:** 94.1% (1755 / 1865)

### simpleqa_nous

| Metric | Score | Std Error |
|--------|-------|----------|
| exact_match | 0.030 | 0.003 |
| fuzzy_match | 0.043 | 0.003 |

**Model:** h4-14b-more-stage1-nonreasoning
**Evaluation Time (hh:mm:ss):** 00:17:00
**Temperature:** 0.6
**Overlong samples:** 100.0% (4321 / 4321)