Add files using upload-large-folder tool

.gitattributes

@@ -33,3 +33,29 @@ 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_10/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+global_step_100/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+global_step_20/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+global_step_30/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+global_step_40/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+global_step_50/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+global_step_60/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
+global_step_90/actor/huggingface/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results/plots/eval_results_acc_keywords.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results/plots/eval_results_acc_pass_acc.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results/plots/eval_results_acc_tokens.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results/plots/eval_results_avg_stop_tokens.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results/plots/eval_results_box_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results/plots/eval_results_clip_ratio.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results/plots/eval_results_correct_tokens.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results/plots/eval_results_repeat_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results/plots/eval_results_tokens_keywords.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results/plots/eval_results_wrong_tokens.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results_avg4/plots/eval_results_avg4_acc_keywords.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results_avg4/plots/eval_results_avg4_acc_pass_acc.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results_avg4/plots/eval_results_avg4_acc_tokens.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results_avg4/plots/eval_results_avg4_avg_stop_tokens.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results_avg4/plots/eval_results_avg4_clip_ratio.png filter=lfs diff=lfs merge=lfs -text
+reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/eval_results_avg4/plots/eval_results_avg4_box_ratio_and_token_length.png filter=lfs diff=lfs merge=lfs -text

global_step_10/actor/huggingface/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

global_step_10/data.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d7c26916a0b5bf53242298f37baf627a9b0ed4e16fabfd78a877ab3cb35472c
+size 1947

global_step_100/actor/huggingface/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

global_step_100/data.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95f3eaab16577c8b15e0e26c7e76ea89028ecdefaee47b9d4ad925c5a562a622
+size 1947

global_step_20/actor/huggingface/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

global_step_20/data.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66ce7c51037d5f729b2a767511d5573796bed9a99711ff83eadb00f227957707
+size 1947

global_step_30/actor/huggingface/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

global_step_30/data.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1dd0fee6dcd750e5c3b43665d8e8da560439ee4d1ed1ac4f0ced642981e02ca1
+size 1947

global_step_40/actor/huggingface/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

global_step_40/data.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7cd793991b9fe00986d0c1d3593ad31fac549dd780b01d5f6d9becd1a8161e69
+size 1947

global_step_50/actor/huggingface/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

global_step_50/data.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c36aed214995f2c4f55495ab6ac7b5ec006bfed7a645a3f139c885d98ba2ed60
+size 1947

global_step_60/actor/huggingface/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

global_step_60/data.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:386f4e1ce32469eb9649ae72e1114453155232915883c775bad93899892df882
+size 1947

global_step_70/actor/huggingface/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

global_step_70/data.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f70670c9ecd962978294e131e8948e9bed7a8e05fd302cd1058ff4e962c7f189
+size 1947

global_step_80/actor/huggingface/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

global_step_80/data.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2da91f2b0ee491fc6c032b29eb667535f053724d53bd035fe62753317a43ae00
+size 1947

global_step_90/actor/huggingface/tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aeb13307a71acd8fe81861d94ad54ab689df773318809eed3cbe794b4492dae4
+size 11422654

global_step_90/data.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4609719c846c982e5c9b89f6f8b8702de5c4ded573a84e527f87e33c095bdbc7
+size 1947

reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/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": 67.5,
+    "pass_acc": 85.0,
+    "pass@k": {
+        "1": 67.5,
+        "2": 76.7,
+        "4": 85.0
+    },
+    "time_use_in_second": 162.94680261611938,
+    "time_use_in_minite": "2:42"
+}

reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/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": 27.5,
+    "pass_acc": 36.7,
+    "pass@k": {
+        "1": 27.5,
+        "2": 32.2,
+        "4": 36.7
+    },
+    "time_use_in_second": 419.3475561141968,
+    "time_use_in_minite": "6:59"
+}

reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/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": 19.2,
+    "pass_acc": 30.0,
+    "pass@k": {
+        "1": 19.2,
+        "2": 25.6,
+        "4": 30.0
+    },
+    "time_use_in_second": 375.0433328151703,
+    "time_use_in_minite": "6:15"
+}

reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/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": 64.4,
+    "pass_acc": 80.0,
+    "pass@k": {
+        "1": 64.4,
+        "2": 74.2,
+        "4": 80.0
+    },
+    "time_use_in_second": 195.60499334335327,
+    "time_use_in_minite": "3:15"
+}

reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/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": 0,
+    "acc": 28.3,
+    "pass_acc": 46.7,
+    "pass@k": {
+        "1": 28.3,
+        "2": 36.1,
+        "4": 46.7
+    },
+    "time_use_in_second": 549.4302027225494,
+    "time_use_in_minite": "9:09"
+}

reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/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": 0,
+    "acc": 22.5,
+    "pass_acc": 36.7,
+    "pass@k": {
+        "1": 22.5,
+        "2": 28.9,
+        "4": 36.7
+    },
+    "time_use_in_second": 381.60382318496704,
+    "time_use_in_minite": "6:21"
+}

reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/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": 0,
+    "acc": 66.9,
+    "pass_acc": 80.0,
+    "pass@k": {
+        "1": 66.9,
+        "2": 73.8,
+        "4": 80.0
+    },
+    "time_use_in_second": 233.23579001426697,
+    "time_use_in_minite": "3:53"
+}

reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/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": 25.0,
+    "pass_acc": 43.3,
+    "pass@k": {
+        "1": 25.0,
+        "2": 32.8,
+        "4": 43.3
+    },
+    "time_use_in_second": 509.6931164264679,
+    "time_use_in_minite": "8:29"
+}

reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/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": 24.2,
+    "pass_acc": 33.3,
+    "pass@k": {
+        "1": 24.2,
+        "2": 28.9,
+        "4": 33.3
+    },
+    "time_use_in_second": 440.93800044059753,
+    "time_use_in_minite": "7:20"
+}

reinforce_pp_deepmath_train_sample_6144_context_4k_Qwen3-8B-Base_max_response4096_batch1024_ppomini256_rollout8_vllm/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": 68.8,
+    "pass_acc": 87.5,
+    "pass@k": {
+        "1": 68.8,
+        "2": 80.8,
+        "4": 87.5
+    },
+    "time_use_in_second": 134.2269902229309,
+    "time_use_in_minite": "2:14"
+}