checkpoint-250

verifier-qwen3.5-2b/{checkpoint-150 → checkpoint-250}/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b4d79dd8ee2636cc3dd1e22cfb16910b13a659d391671204d8e83a90018e3f4c
+oid sha256:fbbb89e4de20fd72514f274abecdf708c6865451b15d904b7203c8c670472c6d
 size 134604152

verifier-qwen3.5-2b/{checkpoint-150 → checkpoint-250}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9a5e893e8bf2e6053ad0d3d3d0a2311845ac9a65f32de7205b9217224f50dc5
+oid sha256:8dd73e75976981f44a79ee67ae425b95fd2368fc919c580b120a52b5db8f407b
 size 269426431

verifier-qwen3.5-2b/{checkpoint-150 → checkpoint-250}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ffe7ec2ad3766a57899742484073683fb6bebb00c616b1d25b54468fc22f2850
+oid sha256:78960911d97e49c97f7fff05fc1c4cbd277b50dfd6917aa7938074deb3e11617
 size 14645

verifier-qwen3.5-2b/{checkpoint-150 → checkpoint-250}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c2893be9da85da3b6ea68754d13bd2c7439b98cbf1e53b420288ad273018452
+oid sha256:3f867c4bc2e54cf7591ecf35a25bfcb415f6f9227a120b27baab0178fa899174
 size 1465

verifier-qwen3.5-2b/{checkpoint-150 → checkpoint-250}/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 150,
-  "best_metric": 0.22001148760318756,
-  "best_model_checkpoint": "./outputs/verifier-qwen3.5-2b/checkpoint-150",
-  "epoch": 0.5181347150259067,
+  "best_global_step": 250,
+  "best_metric": 0.18705631792545319,
+  "best_model_checkpoint": "./outputs/verifier-qwen3.5-2b/checkpoint-250",
+  "epoch": 0.8635578583765112,
   "eval_steps": 50,
-  "global_step": 150,
+  "global_step": 250,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -137,6 +137,92 @@
       "eval_samples_per_second": 118.534,
       "eval_steps_per_second": 15.262,
       "step": 150
+    },
+    {
+      "epoch": 0.5526770293609672,
+      "grad_norm": 1.7403196096420288,
+      "learning_rate": 7.640653357531761e-05,
+      "loss": 0.16722545623779297,
+      "step": 160
+    },
+    {
+      "epoch": 0.5872193436960277,
+      "grad_norm": 1.5578631162643433,
+      "learning_rate": 7.459165154264974e-05,
+      "loss": 0.1962692379951477,
+      "step": 170
+    },
+    {
+      "epoch": 0.6217616580310881,
+      "grad_norm": 1.2932735681533813,
+      "learning_rate": 7.277676950998185e-05,
+      "loss": 0.18478379249572754,
+      "step": 180
+    },
+    {
+      "epoch": 0.6563039723661486,
+      "grad_norm": 1.350063443183899,
+      "learning_rate": 7.096188747731398e-05,
+      "loss": 0.18056931495666503,
+      "step": 190
+    },
+    {
+      "epoch": 0.690846286701209,
+      "grad_norm": 1.4605202674865723,
+      "learning_rate": 6.914700544464611e-05,
+      "loss": 0.19524022340774536,
+      "step": 200
+    },
+    {
+      "epoch": 0.690846286701209,
+      "eval_loss": 0.20766094326972961,
+      "eval_runtime": 1.9685,
+      "eval_samples_per_second": 118.366,
+      "eval_steps_per_second": 15.24,
+      "step": 200
+    },
+    {
+      "epoch": 0.7253886010362695,
+      "grad_norm": 1.6113790273666382,
+      "learning_rate": 6.733212341197823e-05,
+      "loss": 0.19727052450180055,
+      "step": 210
+    },
+    {
+      "epoch": 0.7599309153713298,
+      "grad_norm": 2.703947067260742,
+      "learning_rate": 6.551724137931034e-05,
+      "loss": 0.17285306453704835,
+      "step": 220
+    },
+    {
+      "epoch": 0.7944732297063903,
+      "grad_norm": 2.3148365020751953,
+      "learning_rate": 6.370235934664247e-05,
+      "loss": 0.16258671283721923,
+      "step": 230
+    },
+    {
+      "epoch": 0.8290155440414507,
+      "grad_norm": 1.1692686080932617,
+      "learning_rate": 6.18874773139746e-05,
+      "loss": 0.1559024453163147,
+      "step": 240
+    },
+    {
+      "epoch": 0.8635578583765112,
+      "grad_norm": 2.6930367946624756,
+      "learning_rate": 6.007259528130672e-05,
+      "loss": 0.17801055908203126,
+      "step": 250
+    },
+    {
+      "epoch": 0.8635578583765112,
+      "eval_loss": 0.18705631792545319,
+      "eval_runtime": 1.9703,
+      "eval_samples_per_second": 118.258,
+      "eval_steps_per_second": 15.226,
+      "step": 250
     }
   ],
   "logging_steps": 10,
@@ -156,7 +242,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4732870507524096.0,
+  "total_flos": 7895230819811328.0,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

verifier-qwen3.5-2b/tb/events.out.tfevents.1790667609.02b48645612f.2184.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bb1a368d629d0a116e90b00ab92ccd6a23c7e08074d897f352f0eebf3db850c2
-size 10679
+oid sha256:bcf8d2eca3e02ca82dd1a427e475acee33320ad1bb26d5403d8135ef72edfb37
+size 12005