checkpoint-300

verifier-qwen3.5-2b/{checkpoint-200 → checkpoint-300}/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bf59e7241b6d9f9d5521691a4dbb76d67ca51d0facb7c81f605a0471a6ecf570
+oid sha256:b4fa7ae17b2f05b7af9e449e561e0197025f5b067ee186b263e59370fac50df3
 size 134604152

verifier-qwen3.5-2b/{checkpoint-200 → checkpoint-300}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f5b725bc74a325a963e4de2c61eb69793de9a80c5bd4ca2b705f7493b6e0a694
+oid sha256:e1b375d1d90416d333adb8eae939dd2a9b0bd62ec4f08f987f35a3f67f8abeb9
 size 269426431

verifier-qwen3.5-2b/{checkpoint-200 → checkpoint-300}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f9197cb2325b9b47e226057dae57f81670041e1727b4f5c0693aca5457b601e
+oid sha256:45acf99c56b55aba32fc5d6dd3792f86d57ad4b9d765e544f39d9a717f32c9c3
 size 14645

verifier-qwen3.5-2b/{checkpoint-200 → checkpoint-300}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d39e1efc00e6fe7a440b3b47997a336e3ba2211044f8a50e0a9d6394db6f64c
+oid sha256:a77c0652392301021f9fa41b6c909f0f649f2c1dc97aa0e1404793fdc2d3e77e
 size 1465

verifier-qwen3.5-2b/{checkpoint-200 → checkpoint-300}/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 200,
-  "best_metric": 0.20766094326972961,
-  "best_model_checkpoint": "./outputs/verifier-qwen3.5-2b/checkpoint-200",
-  "epoch": 0.690846286701209,
+  "best_global_step": 250,
+  "best_metric": 0.18705631792545319,
+  "best_model_checkpoint": "./outputs/verifier-qwen3.5-2b/checkpoint-250",
+  "epoch": 1.0345423143350605,
   "eval_steps": 50,
-  "global_step": 200,
+  "global_step": 300,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -180,6 +180,92 @@
       "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
+    },
+    {
+      "epoch": 0.8981001727115717,
+      "grad_norm": 1.8818975687026978,
+      "learning_rate": 5.825771324863884e-05,
+      "loss": 0.19102323055267334,
+      "step": 260
+    },
+    {
+      "epoch": 0.9326424870466321,
+      "grad_norm": 1.780072808265686,
+      "learning_rate": 5.644283121597096e-05,
+      "loss": 0.15900615453720093,
+      "step": 270
+    },
+    {
+      "epoch": 0.9671848013816926,
+      "grad_norm": 2.9257590770721436,
+      "learning_rate": 5.4627949183303086e-05,
+      "loss": 0.14064459800720214,
+      "step": 280
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 4.117520809173584,
+      "learning_rate": 5.281306715063521e-05,
+      "loss": 0.18647940158843995,
+      "step": 290
+    },
+    {
+      "epoch": 1.0345423143350605,
+      "grad_norm": 0.9831042885780334,
+      "learning_rate": 5.099818511796734e-05,
+      "loss": 0.10912895202636719,
+      "step": 300
+    },
+    {
+      "epoch": 1.0345423143350605,
+      "eval_loss": 0.2040945291519165,
+      "eval_runtime": 1.9743,
+      "eval_samples_per_second": 118.017,
+      "eval_steps_per_second": 15.195,
+      "step": 300
     }
   ],
   "logging_steps": 10,
@@ -199,7 +285,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6313645475278848.0,
+  "total_flos": 9473439594436608.0,
   "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:bcf8d2eca3e02ca82dd1a427e475acee33320ad1bb26d5403d8135ef72edfb37
-size 12005
+oid sha256:81842c697a26ec113f61b611cf6af004701819dba73d9b4691dd418f04619931
+size 13331