eval-Cogito-v2-preview-70B-nonreasoning

# cogito-70b-nonthinking Evaluation Results

## Summary

| Benchmark | Score | Metric | Samples | Overlong rate |
|-----------|-------|--------|---------|---------------|
| aime24 | 0.122 | math_pass@1:64_samples | 64 | 100.0% |
| aime25 | 0.060 | math_pass@1:64_samples | 64 | 100.0% |
| arenahard | 0.819 | eval/overall_winrate | 500 | 0.0% |
| bbh_generative | 0.876 | extractive_match | 1 | 100.0% |
| creative-writing-v3 | 0.655 | creative_writing_score | 96 | 0.0% |
| drop_generative_nous | 0.841 | drop_acc | 1 | 100.0% |
| eqbench3 | 0.681 | eqbench_score | 135 | 0.0% |
| gpqa_diamond | 0.528 | gpqa_pass@1:8_samples | 8 | 100.0% |
| ifeval | 0.927 | inst_level_loose_acc | 1 | 100.0% |
| lcb-v6-aug2024+ | 0.272 | eval/pass_1 | 1 | 100.0% |
| math_500 | 0.756 | math_pass@1:4_samples | 4 | 100.0% |
| mmlu_generative | 0.905 | extractive_match | 1 | 100.0% |
| mmlu_pro | 0.760 | pass@1:1_samples | 1 | 100.0% |
| musr_generative | 0.592 | extractive_match | 1 | 100.0% |
| obqa_generative | 0.942 | extractive_match | 1 | 100.0% |
| rewardbench | 0.627 | eval/percent_correct | 1 | 94.5% |
| simpleqa_nous | 0.227 | fuzzy_match | 1 | 100.0% |

Overlong rate: 63,690 / 64,523 samples (98.7%) missing closing `</think>` tag

## Detailed Results

### aime24

| Metric | Score | Std Error |
|--------|-------|----------|
| math_pass@1:1_samples | 0.133 | 0.063 |
| math_pass@1:4_samples | 0.117 | 0.052 |
| math_pass@1:8_samples | 0.117 | 0.049 |
| math_pass@1:16_samples | 0.115 | 0.048 |
| math_pass@1:32_samples | 0.118 | 0.048 |
| math_pass@1:64_samples | 0.122 | 0.048 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:19:49
**Temperature:** 0.6
**Overlong samples:** 100.0% (1920 / 1920)

### aime25

| Metric | Score | Std Error |
|--------|-------|----------|
| math_pass@1:1_samples | 0.067 | 0.046 |
| math_pass@1:4_samples | 0.058 | 0.037 |
| math_pass@1:8_samples | 0.058 | 0.031 |
| math_pass@1:16_samples | 0.058 | 0.029 |
| math_pass@1:32_samples | 0.067 | 0.030 |
| math_pass@1:64_samples | 0.060 | 0.029 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:17:57
**Temperature:** 0.6
**Overlong samples:** 100.0% (1920 / 1920)

### arenahard

| Metric | Score | Std Error |
|--------|-------|----------|
| eval/overall_winrate | 0.819 | 0.000 |
| eval/total_samples | 500.000 | 0.000 |
| eval/win_count | 372.000 | 0.000 |
| eval/tie_count | 74.000 | 0.000 |
| eval/loss_count | 54.000 | 0.000 |
| eval/win_rate | 0.744 | 0.000 |
| eval/tie_rate | 0.148 | 0.000 |
| eval/loss_rate | 0.108 | 0.000 |
| eval/winrate_arena-hard-v0.1 | 0.819 | 0.000 |

**Model:** cogito-70b-arena-nothink
**Evaluation Time (hh:mm:ss):** 00:02:07
**Temperature:** 0.6
**Overlong samples:** 0.0% (0 / 500)

### bbh_generative

| Metric | Score | Std Error |
|--------|-------|----------|
| extractive_match | 0.876 | 0.015 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:24:48
**Temperature:** 0.6
**Overlong samples:** 100.0% (5511 / 5511)

### creative-writing-v3

| Metric | Score | Std Error |
|--------|-------|----------|
| creative_writing_score | 0.655 | 0.000 |
| num_samples | 96.000 | 0.000 |

**Model:** cogito-70b-nonthinking
**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 | 0.841 | 0.004 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:37:27
**Temperature:** 0.6
**Overlong samples:** 100.0% (9536 / 9536)

### eqbench3

| Metric | Score | Std Error |
|--------|-------|----------|
| eqbench_score | 0.681 | 0.000 |
| num_samples | 135.000 | 0.000 |

**Model:** cogito-70b-arena-nothink
**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 | 0.571 | 0.035 |
| gpqa_pass@1:4_samples | 0.529 | 0.027 |
| gpqa_pass@1:8_samples | 0.528 | 0.026 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:15:49
**Temperature:** 0.6
**Overlong samples:** 100.0% (1584 / 1584)

### ifeval

| Metric | Score | Std Error |
|--------|-------|----------|
| prompt_level_strict_acc | 0.865 | 0.015 |
| inst_level_strict_acc | 0.911 | 0.000 |
| prompt_level_loose_acc | 0.889 | 0.014 |
| inst_level_loose_acc | 0.927 | 0.000 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:02:18
**Temperature:** 0.6
**Overlong samples:** 100.0% (541 / 541)

### lcb-v6-aug2024+

| Metric | Score | Std Error |
|--------|-------|----------|
| eval/pass_1 | 0.272 | 0.000 |
| eval/easy_pass_1 | 0.751 | 0.000 |
| eval/medium_pass_1 | 0.212 | 0.000 |
| eval/hard_pass_1 | 0.055 | 0.000 |
| eval/completion_length | 1847.961 | 0.000 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:18:42
**Temperature:** N/A
**Overlong samples:** 100.0% (7264 / 7264)

### math_500

| Metric | Score | Std Error |
|--------|-------|----------|
| math_pass@1:1_samples | 0.750 | 0.019 |
| math_pass@1:4_samples | 0.756 | 0.016 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:12:54
**Temperature:** 0.6
**Overlong samples:** 100.0% (2000 / 2000)

### mmlu_generative

| Metric | Score | Std Error |
|--------|-------|----------|
| extractive_match | 0.905 | 0.002 |

**Model:** cogito-70b-nonthinking
**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 | 0.760 | 0.004 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 01:01:21
**Temperature:** 0.6
**Overlong samples:** 100.0% (12032 / 12032)

### musr_generative

| Metric | Score | Std Error |
|--------|-------|----------|
| extractive_match | 0.592 | 0.031 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:03:01
**Temperature:** 0.6
**Overlong samples:** 100.0% (756 / 756)

### obqa_generative

| Metric | Score | Std Error |
|--------|-------|----------|
| extractive_match | 0.942 | 0.010 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:02:01
**Temperature:** 0.6
**Overlong samples:** 100.0% (500 / 500)

### rewardbench

| Metric | Score | Std Error |
|--------|-------|----------|
| eval/percent_correct | 0.627 | 0.000 |
| eval/total_samples | 1865.000 | 0.000 |
| eval/correct_samples | 1170.000 | 0.000 |
| eval/format_compliance_rate | 1.000 | 0.000 |
| eval/avg_response_length | 1283.707 | 0.000 |
| eval/response_length_std | 209.582 | 0.000 |
| eval/judgment_entropy | 1.367 | 0.000 |
| eval/most_common_judgment_freq | 0.330 | 0.000 |
| eval/format_error_rate | 0.000 | 0.000 |
| eval/avg_ties_rating | 3.614 | 0.000 |
| eval/ties_error_rate | 0.019 | 0.000 |
| eval/percent_correct_Factuality | 0.514 | 0.000 |
| eval/percent_correct_Precise IF | 0.362 | 0.000 |
| eval/percent_correct_Math | 0.497 | 0.000 |
| eval/percent_correct_Safety | 0.627 | 0.000 |
| eval/percent_correct_Focus | 0.804 | 0.000 |
| eval/percent_correct_Ties | 0.951 | 0.000 |
| eval/choice_samples | 1763.000 | 0.000 |
| eval/ties_samples | 102.000 | 0.000 |
| eval/choice_format_compliance_rate | 1.000 | 0.000 |
| eval/ties_format_compliance_rate | 1.000 | 0.000 |
| eval/wrong_answer_a_bias_rate | 0.371 | 0.000 |
| eval/wrong_answer_total_count | 690.000 | 0.000 |
| eval/wrong_answer_a_count | 256.000 | 0.000 |

**Model:** cogito-70b-arena-nothink
**Evaluation Time (hh:mm:ss):** 00:05:11
**Temperature:** 0.6
**Overlong samples:** 94.5% (1763 / 1865)

### simpleqa_nous

| Metric | Score | Std Error |
|--------|-------|----------|
| exact_match | 0.167 | 0.006 |
| fuzzy_match | 0.227 | 0.006 |

**Model:** cogito-70b-nonthinking
**Evaluation Time (hh:mm:ss):** 00:16:58
**Temperature:** 0.6
**Overlong samples:** 100.0% (4321 / 4321)

# cogito-70b-nonthinking 评测结果

## 摘要
| 评测基准 | 得分 | 评测指标 | 样本量 | 超长样本占比 |
|-----------|-------|--------|---------|---------------|
| aime24 | 0.122 | math_pass@1:64_samples | 64 | 100.0% |
| aime25 | 0.060 | math_pass@1:64_samples | 64 | 100.0% |
| arenahard | 0.819 | eval/overall_winrate | 500 | 0.0% |
| bbh_generative | 0.876 | extractive_match | 1 | 100.0% |
| creative-writing-v3 | 0.655 | creative_writing_score | 96 | 0.0% |
| drop_generative_nous | 0.841 | drop_acc | 1 | 100.0% |
| eqbench3 | 0.681 | eqbench_score | 135 | 0.0% |
| gpqa_diamond | 0.528 | gpqa_pass@1:8_samples | 8 | 100.0% |
| ifeval | 0.927 | inst_level_loose_acc | 1 | 100.0% |
| lcb-v6-aug2024+ | 0.272 | eval/pass_1 | 1 | 100.0% |
| math_500 | 0.756 | math_pass@1:4_samples | 4 | 100.0% |
| mmlu_generative | 0.905 | extractive_match | 1 | 100.0% |
| mmlu_pro | 0.760 | pass@1:1_samples | 1 | 100.0% |
| musr_generative | 0.592 | extractive_match | 1 | 100.0% |
| obqa_generative | 0.942 | extractive_match | 1 | 100.0% |
| rewardbench | 0.627 | eval/percent_correct | 1 | 94.5% |
| simpleqa_nous | 0.227 | fuzzy_match | 1 | 100.0% |

整体超长样本情况：63,690 / 64,523 个样本（占比98.7%）缺失闭合标签`</think>`

## 详细评测结果

### aime24
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| math_pass@1:1_samples | 0.133 | 0.063 |
| math_pass@1:4_samples | 0.117 | 0.052 |
| math_pass@1:8_samples | 0.117 | 0.049 |
| math_pass@1:16_samples | 0.115 | 0.048 |
| math_pass@1:32_samples | 0.118 | 0.048 |
| math_pass@1:64_samples | 0.122 | 0.048 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:19:49
**温度参数：** 0.6
**超长样本情况：** 100.0% (1920 / 1920)

### aime25
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| math_pass@1:1_samples | 0.067 | 0.046 |
| math_pass@1:4_samples | 0.058 | 0.037 |
| math_pass@1:8_samples | 0.058 | 0.031 |
| math_pass@1:16_samples | 0.058 | 0.029 |
| math_pass@1:32_samples | 0.067 | 0.030 |
| math_pass@1:64_samples | 0.060 | 0.029 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:17:57
**温度参数：** 0.6
**超长样本情况：** 100.0% (1920 / 1920)

### arenahard
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| eval/overall_winrate | 0.819 | 0.000 |
| eval/total_samples | 500.000 | 0.000 |
| eval/win_count | 372.000 | 0.000 |
| eval/tie_count | 74.000 | 0.000 |
| eval/loss_count | 54.000 | 0.000 |
| eval/win_rate | 0.744 | 0.000 |
| eval/tie_rate | 0.148 | 0.000 |
| eval/loss_rate | 0.108 | 0.000 |
| eval/winrate_arena-hard-v0.1 | 0.819 | 0.000 |

**模型：** cogito-70b-arena-nothink
**评测耗时（时:分:秒）：** 00:02:07
**温度参数：** 0.6
**超长样本情况：** 0.0% (0 / 500)

### bbh_generative
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| extractive_match | 0.876 | 0.015 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:24:48
**温度参数：** 0.6
**超长样本情况：** 100.0% (5511 / 5511)

### creative-writing-v3
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| creative_writing_score | 0.655 | 0.000 |
| num_samples | 96.000 | 0.000 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** N/A
**温度参数：** N/A
**超长样本情况：** 0.0% (0 / 96)

### drop_generative_nous
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| drop_acc | 0.841 | 0.004 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:37:27
**温度参数：** 0.6
**超长样本情况：** 100.0% (9536 / 9536)

### eqbench3
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| eqbench_score | 0.681 | 0.000 |
| num_samples | 135.000 | 0.000 |

**模型：** cogito-70b-arena-nothink
**评测耗时（时:分:秒）：** N/A
**温度参数：** N/A
**超长样本情况：** 0.0% (0 / 135)

### gpqa_diamond
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| gpqa_pass@1:1_samples | 0.571 | 0.035 |
| gpqa_pass@1:4_samples | 0.529 | 0.027 |
| gpqa_pass@1:8_samples | 0.528 | 0.026 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:15:49
**温度参数：** 0.6
**超长样本情况：** 100.0% (1584 / 1584)

### ifeval
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| prompt_level_strict_acc | 0.865 | 0.015 |
| inst_level_strict_acc | 0.911 | 0.000 |
| prompt_level_loose_acc | 0.889 | 0.014 |
| inst_level_loose_acc | 0.927 | 0.000 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:02:18
**温度参数：** 0.6
**超长样本情况：** 100.0% (541 / 541)

### lcb-v6-aug2024+
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| eval/pass_1 | 0.272 | 0.000 |
| eval/easy_pass_1 | 0.751 | 0.000 |
| eval/medium_pass_1 | 0.212 | 0.000 |
| eval/hard_pass_1 | 0.055 | 0.000 |
| eval/completion_length | 1847.961 | 0.000 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:18:42
**温度参数：** N/A
**超长样本情况：** 100.0% (7264 / 7264)

### math_500
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| math_pass@1:1_samples | 0.750 | 0.019 |
| math_pass@1:4_samples | 0.756 | 0.016 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:12:54
**温度参数：** 0.6
**超长样本情况：** 100.0% (2000 / 2000)

### mmlu_generative
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| extractive_match | 0.905 | 0.002 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:55:09
**温度参数：** 0.6
**超长样本情况：** 100.0% (14042 / 14042)

### mmlu_pro
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| pass@1:1_samples | 0.760 | 0.004 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 01:01:21
**温度参数：** 0.6
**超长样本情况：** 100.0% (12032 / 12032)

### musr_generative
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| extractive_match | 0.592 | 0.031 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:03:01
**温度参数：** 0.6
**超长样本情况：** 100.0% (756 / 756)

### obqa_generative
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| extractive_match | 0.942 | 0.010 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:02:01
**温度参数：** 0.6
**超长样本情况：** 100.0% (500 / 500)

### rewardbench
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| eval/percent_correct | 0.627 | 0.000 |
| eval/total_samples | 1865.000 | 0.000 |
| eval/correct_samples | 1170.000 | 0.000 |
| eval/format_compliance_rate | 1.000 | 0.000 |
| eval/avg_response_length | 1283.707 | 0.000 |
| eval/response_length_std | 209.582 | 0.000 |
| eval/judgment_entropy | 1.367 | 0.000 |
| eval/most_common_judgment_freq | 0.330 | 0.000 |
| eval/format_error_rate | 0.000 | 0.000 |
| eval/avg_ties_rating | 3.614 | 0.000 |
| eval/ties_error_rate | 0.019 | 0.000 |
| eval/percent_correct_Factuality | 0.514 | 0.000 |
| eval/percent_correct_Precise IF | 0.362 | 0.000 |
| eval/percent_correct_Math | 0.497 | 0.000 |
| eval/percent_correct_Safety | 0.627 | 0.000 |
| eval/percent_correct_Focus | 0.804 | 0.000 |
| eval/percent_correct_Ties | 0.951 | 0.000 |
| eval/choice_samples | 1763.000 | 0.000 |
| eval/ties_samples | 102.000 | 0.000 |
| eval/choice_format_compliance_rate | 1.000 | 0.000 |
| eval/ties_format_compliance_rate | 1.000 | 0.000 |
| eval/wrong_answer_a_bias_rate | 0.371 | 0.000 |
| eval/wrong_answer_total_count | 690.000 | 0.000 |
| eval/wrong_answer_a_count | 256.000 | 0.000 |

**模型：** cogito-70b-arena-nothink
**评测耗时（时:分:秒）：** 00:05:11
**温度参数：** 0.6
**超长样本情况：** 94.5% (1763 / 1865)

### simpleqa_nous
| 评测指标 | 得分 | 标准误差 |
|--------|-------|----------|
| exact_match | 0.167 | 0.006 |
| fuzzy_match | 0.227 | 0.006 |

**模型：** cogito-70b-nonthinking
**评测耗时（时:分:秒）：** 00:16:58
**温度参数：** 0.6
**超长样本情况：** 100.0% (4321 / 4321)