Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,25 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+global_step_90/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+global_step_70/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+global_step_80/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+eval_results/global_step_0/mmlu_stem/test_qwen-boxed_-1_seed0_t1.0_s0_e-1.jsonl filter=lfs diff=lfs merge=lfs -text
+eval_results/plots/eval_results_acc_pass_acc.png filter=lfs diff=lfs merge=lfs -text
+eval_results/plots/eval_results_acc_keywords.png filter=lfs diff=lfs merge=lfs -text
+eval_results/plots/eval_results_acc_tokens.png filter=lfs diff=lfs merge=lfs -text
+eval_results/plots/eval_results_avg_stop_tokens.png filter=lfs diff=lfs merge=lfs -text
+eval_results/plots/eval_results_box_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
+eval_results/plots/eval_results_clip_ratio.png filter=lfs diff=lfs merge=lfs -text
+eval_results/plots/eval_results_correct_tokens.png filter=lfs diff=lfs merge=lfs -text
+eval_results/plots/eval_results_repeat_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
+eval_results/plots/eval_results_tokens_keywords.png filter=lfs diff=lfs merge=lfs -text
+eval_results/plots/eval_results_wrong_tokens.png filter=lfs diff=lfs merge=lfs -text
+eval_results_avg4/plots/eval_results_avg4_acc_keywords.png filter=lfs diff=lfs merge=lfs -text
+eval_results_avg4/plots/eval_results_avg4_acc_pass_acc.png filter=lfs diff=lfs merge=lfs -text
+eval_results_avg4/plots/eval_results_avg4_acc_tokens.png filter=lfs diff=lfs merge=lfs -text
+eval_results_avg4/plots/eval_results_avg4_avg_stop_tokens.png filter=lfs diff=lfs merge=lfs -text
+eval_results_avg4/plots/eval_results_avg4_clip_ratio.png filter=lfs diff=lfs merge=lfs -text
+eval_results_avg4/plots/eval_results_avg4_box_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
+eval_results_avg4/plots/eval_results_avg4_correct_tokens.png filter=lfs diff=lfs merge=lfs -text
+eval_results_avg4/plots/eval_results_avg4_tokens_keywords.png filter=lfs diff=lfs merge=lfs -text

eval_results/global_step_0/mmlu_stem/test_qwen-boxed_-1_seed0_t1.0_s0_e-1.jsonl

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a42fa1343b5e1c8505957614a09bf3ab4ed175d5ab2b1c05f2ab9dc1ba0db40
+size 12132789

eval_results_avg4/global_step_50/amc23/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 40,
+    "num_scores": 160,
+    "timeout_samples": 0,
+    "empty_samples": 0,
+    "acc": 40.0,
+    "pass_acc": 60.0,
+    "pass@k": {
+        "1": 40.0,
+        "2": 50.4,
+        "4": 60.0
+    },
+    "time_use_in_second": 21.898350715637207,
+    "time_use_in_minite": "0:21"
+}

eval_results_avg4/global_step_60/aime24/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 30,
+    "num_scores": 120,
+    "timeout_samples": 0,
+    "empty_samples": 0,
+    "acc": 4.2,
+    "pass_acc": 6.7,
+    "pass@k": {
+        "1": 4.2,
+        "2": 5.0,
+        "4": 6.7
+    },
+    "time_use_in_second": 20.911217212677002,
+    "time_use_in_minite": "0:20"
+}

eval_results_avg4/global_step_60/aime25/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 30,
+    "num_scores": 120,
+    "timeout_samples": 0,
+    "empty_samples": 0,
+    "acc": 1.7,
+    "pass_acc": 3.3,
+    "pass@k": {
+        "1": 1.7,
+        "2": 2.8,
+        "4": 3.3
+    },
+    "time_use_in_second": 18.055476665496826,
+    "time_use_in_minite": "0:18"
+}

eval_results_avg4/global_step_60/amc23/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 40,
+    "num_scores": 160,
+    "timeout_samples": 0,
+    "empty_samples": 0,
+    "acc": 44.4,
+    "pass_acc": 62.5,
+    "pass@k": {
+        "1": 44.4,
+        "2": 56.7,
+        "4": 62.5
+    },
+    "time_use_in_second": 21.025193214416504,
+    "time_use_in_minite": "0:21"
+}

eval_results_avg4/global_step_70/aime24/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 30,
+    "num_scores": 120,
+    "timeout_samples": 0,
+    "empty_samples": 0,
+    "acc": 9.2,
+    "pass_acc": 23.3,
+    "pass@k": {
+        "1": 9.2,
+        "2": 14.4,
+        "4": 23.3
+    },
+    "time_use_in_second": 20.478790760040283,
+    "time_use_in_minite": "0:20"
+}

eval_results_avg4/global_step_70/aime25/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 30,
+    "num_scores": 120,
+    "timeout_samples": 0,
+    "empty_samples": 0,
+    "acc": 0.8,
+    "pass_acc": 3.3,
+    "pass@k": {
+        "1": 0.8,
+        "2": 1.7,
+        "4": 3.3
+    },
+    "time_use_in_second": 19.925731897354126,
+    "time_use_in_minite": "0:19"
+}

eval_results_avg4/global_step_70/amc23/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 40,
+    "num_scores": 160,
+    "timeout_samples": 0,
+    "empty_samples": 0,
+    "acc": 35.0,
+    "pass_acc": 60.0,
+    "pass@k": {
+        "1": 35.0,
+        "2": 46.2,
+        "4": 60.0
+    },
+    "time_use_in_second": 22.096985340118408,
+    "time_use_in_minite": "0:22"
+}

eval_results_avg4/global_step_80/aime24/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 30,
+    "num_scores": 120,
+    "timeout_samples": 0,
+    "empty_samples": 1,
+    "acc": 7.5,
+    "pass_acc": 13.3,
+    "pass@k": {
+        "1": 7.5,
+        "2": 10.6,
+        "4": 13.3
+    },
+    "time_use_in_second": 20.141565561294556,
+    "time_use_in_minite": "0:20"
+}

eval_results_avg4/global_step_80/aime25/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 30,
+    "num_scores": 120,
+    "timeout_samples": 0,
+    "empty_samples": 1,
+    "acc": 2.5,
+    "pass_acc": 6.7,
+    "pass@k": {
+        "1": 2.5,
+        "2": 4.4,
+        "4": 6.7
+    },
+    "time_use_in_second": 18.300720691680908,
+    "time_use_in_minite": "0:18"
+}

eval_results_avg4/global_step_80/amc23/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 40,
+    "num_scores": 160,
+    "timeout_samples": 0,
+    "empty_samples": 1,
+    "acc": 30.6,
+    "pass_acc": 55.0,
+    "pass@k": {
+        "1": 30.6,
+        "2": 42.9,
+        "4": 55.0
+    },
+    "time_use_in_second": 19.80013918876648,
+    "time_use_in_minite": "0:19"
+}

eval_results_avg4/global_step_90/aime24/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 30,
+    "num_scores": 120,
+    "timeout_samples": 0,
+    "empty_samples": 0,
+    "acc": 4.2,
+    "pass_acc": 10.0,
+    "pass@k": {
+        "1": 4.2,
+        "2": 6.7,
+        "4": 10.0
+    },
+    "time_use_in_second": 20.099088191986084,
+    "time_use_in_minite": "0:20"
+}

eval_results_avg4/global_step_90/aime25/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 30,
+    "num_scores": 120,
+    "timeout_samples": 0,
+    "empty_samples": 0,
+    "acc": 0.0,
+    "pass_acc": 0.0,
+    "pass@k": {
+        "1": 0.0,
+        "2": 0.0,
+        "4": 0.0
+    },
+    "time_use_in_second": 16.414506196975708,
+    "time_use_in_minite": "0:16"
+}

eval_results_avg4/global_step_90/amc23/test_qwen-boxed_-1_seed0_t1.0_s0_e-1_metrics.json

@@ -0,0 +1,15 @@
+{
+    "num_samples": 40,
+    "num_scores": 160,
+    "timeout_samples": 0,
+    "empty_samples": 0,
+    "acc": 26.2,
+    "pass_acc": 40.0,
+    "pass@k": {
+        "1": 26.2,
+        "2": 33.3,
+        "4": 40.0
+    },
+    "time_use_in_second": 20.395983457565308,
+    "time_use_in_minite": "0:20"
+}

eval_results_merged/merged.csv

@@ -0,0 +1,12 @@
+global_step,aime24_acc_avg4,aime25_acc_avg4,amc23_acc_avg4,aime24_acc,aime25_acc,amc23_acc,gsm8k_acc,math500_acc,minerva_math_acc,olympiadbench_acc,mmlu_stem_acc,prompt_level_strict_acc_ood,gpqa_pass@1:1_samples_ood
+0,3.30,2.50,25.00,3.30,3.30,22.50,59.40,41.00,11.00,19.30,39.10,20.5,
+10,4.20,3.30,32.50,6.70,6.70,37.50,74.10,57.80,21.00,21.00,42.90,22.0,
+20,5.00,2.50,35.60,6.70,0.00,25.00,77.60,59.40,24.60,24.10,46.00,25.9,
+30,5.00,5.80,36.20,6.70,10.00,35.00,78.90,63.20,26.80,24.90,50.60,29.2,
+40,5.80,1.70,35.00,10.00,3.30,40.00,80.90,64.00,26.10,28.40,56.80,27.7,
+50,5.80,0.80,40.00,3.30,3.30,35.00,81.00,63.20,27.90,27.00,57.70,25.7,
+60,4.20,1.70,44.40,6.70,3.30,22.50,80.20,64.20,24.60,25.90,59.20,26.1,
+70,9.20,0.80,35.00,13.30,6.70,32.50,80.70,63.20,27.20,26.10,60.90,25.5,
+80,7.50,2.50,30.60,3.30,0.00,25.00,78.70,57.00,21.70,22.70,44.60,22.4,
+90,4.20,0.00,26.20,3.30,3.30,37.50,80.00,60.40,21.70,25.30,53.10,24.0,
+100,4.20,0.00,30.00,6.70,0.00,30.00,80.90,56.20,21.00,25.30,45.30,22.2,