checkpoint-250

verifier-qwen3.5-2b-v2/{checkpoint-150 → checkpoint-250}/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5119d4731772c5145fad91b568e905541e8ed9df099fecbac4785ad6da3c39b3
+oid sha256:cda9d44edb27c1448f8fa7efbe32b427dfd30be8de47d7c99b7d7bf2abfcf290
 size 134604152

verifier-qwen3.5-2b-v2/{checkpoint-150 → checkpoint-250}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:368c452a48fb05f910ea28c0d41ec2523b077e8b814887b52f3225a2f57a0730
+oid sha256:354bf122327a67a59f60abf0a22c9a7deba2d0fba760b27becf78238898d2c78
 size 269426431

verifier-qwen3.5-2b-v2/{checkpoint-150 → checkpoint-250}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a0d62c02f66851c8e8ce8922a9abf2545edcbb8b375903ed73ed89debf4523f6
+oid sha256:c404ae259eb57de3586e899e5b05612a0bd2bc6d8b810ef3d0ecfc86791bce34
 size 14645

verifier-qwen3.5-2b-v2/{checkpoint-150 → checkpoint-250}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d3d68fe671ba2b0900dd318a2837b2853470dd977e64be3690ff3f413c9aa1d7
+oid sha256:76981bcdcca847c76a99a657b4fa0a7d19de24aaac18df96a81d8852b2293a3f
 size 1465

verifier-qwen3.5-2b-v2/{checkpoint-150 → checkpoint-250}/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 150,
-  "best_metric": 0.19720077514648438,
-  "best_model_checkpoint": "./outputs/verifier-qwen3.5-2b-v2/checkpoint-150",
-  "epoch": 0.5484460694698354,
+  "best_global_step": 250,
+  "best_metric": 0.19475513696670532,
+  "best_model_checkpoint": "./outputs/verifier-qwen3.5-2b-v2/checkpoint-250",
+  "epoch": 0.9140767824497258,
   "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": 91.046,
       "eval_steps_per_second": 11.748,
       "step": 150
+    },
+    {
+      "epoch": 0.5850091407678245,
+      "grad_norm": 1.8531203269958496,
+      "learning_rate": 7.480769230769231e-05,
+      "loss": 0.19025720357894899,
+      "step": 160
+    },
+    {
+      "epoch": 0.6215722120658135,
+      "grad_norm": 3.6823484897613525,
+      "learning_rate": 7.288461538461538e-05,
+      "loss": 0.2013117790222168,
+      "step": 170
+    },
+    {
+      "epoch": 0.6581352833638026,
+      "grad_norm": 1.0100839138031006,
+      "learning_rate": 7.096153846153847e-05,
+      "loss": 0.15455222129821777,
+      "step": 180
+    },
+    {
+      "epoch": 0.6946983546617916,
+      "grad_norm": 3.111809015274048,
+      "learning_rate": 6.903846153846154e-05,
+      "loss": 0.1986108660697937,
+      "step": 190
+    },
+    {
+      "epoch": 0.7312614259597806,
+      "grad_norm": 4.505012512207031,
+      "learning_rate": 6.711538461538461e-05,
+      "loss": 0.18796275854110717,
+      "step": 200
+    },
+    {
+      "epoch": 0.7312614259597806,
+      "eval_loss": 0.24587298929691315,
+      "eval_runtime": 2.3835,
+      "eval_samples_per_second": 91.042,
+      "eval_steps_per_second": 11.747,
+      "step": 200
+    },
+    {
+      "epoch": 0.7678244972577697,
+      "grad_norm": 1.8294121026992798,
+      "learning_rate": 6.519230769230769e-05,
+      "loss": 0.18504512310028076,
+      "step": 210
+    },
+    {
+      "epoch": 0.8043875685557587,
+      "grad_norm": 2.2052628993988037,
+      "learning_rate": 6.326923076923077e-05,
+      "loss": 0.18328071832656861,
+      "step": 220
+    },
+    {
+      "epoch": 0.8409506398537477,
+      "grad_norm": 2.145005226135254,
+      "learning_rate": 6.134615384615385e-05,
+      "loss": 0.16595417261123657,
+      "step": 230
+    },
+    {
+      "epoch": 0.8775137111517367,
+      "grad_norm": 1.1548988819122314,
+      "learning_rate": 5.942307692307693e-05,
+      "loss": 0.15747179985046386,
+      "step": 240
+    },
+    {
+      "epoch": 0.9140767824497258,
+      "grad_norm": 1.9555728435516357,
+      "learning_rate": 5.7499999999999995e-05,
+      "loss": 0.19622230529785156,
+      "step": 250
+    },
+    {
+      "epoch": 0.9140767824497258,
+      "eval_loss": 0.19475513696670532,
+      "eval_runtime": 2.3841,
+      "eval_samples_per_second": 91.019,
+      "eval_steps_per_second": 11.744,
+      "step": 250
     }
   ],
   "logging_steps": 10,
@@ -156,7 +242,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 4740028835727360.0,
+  "total_flos": 7893880191848448.0,
   "train_batch_size": 8,
   "trial_name": null,
   "trial_params": null

verifier-qwen3.5-2b-v2/tb/events.out.tfevents.1790685928.073545c29144.1706.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d799b7e149ac948bf7e29e98cc328c71a480d5ecbcf9b6c77b5f59d823622f57
-size 10682
+oid sha256:a13f30fa83f5e89c878fd30afea74be1249fb2b01cdb678a0cfeca9884336db4
+size 12008