checkpoint-300

verifier-qwen3.5-2b-v2/{checkpoint-200 → checkpoint-300}/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f64d0d61ae7227cd0a4930a430d1c31fd05a3437c5d4a263e04106d2ac27179
+oid sha256:4ff0f884065cc5ec31f5a4d3fca6560160fdc7f1a18fa2506b9a47625e6bd49a
 size 134604152

verifier-qwen3.5-2b-v2/{checkpoint-200 → checkpoint-300}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2798f5617a6babee74f7dd725c87d5c32580bb157447fb4b7e070022175ac321
+oid sha256:2b1438a7faff6db15b80057654c004d17491265bdc71463b6b5ebf4f24c0fd3f
 size 269426431

verifier-qwen3.5-2b-v2/{checkpoint-200 → checkpoint-300}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:214c39d274aae7a84abefce418c08173a922fa2de0e1a2bad24048cdbd4c2692
+oid sha256:7ea45f0e92806ef13eab47c6d9285ea62305ba2f674eed77dbe12e01824f89d4
 size 14645

verifier-qwen3.5-2b-v2/{checkpoint-200 → checkpoint-300}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9345529b6e002b72948f483d9aa5bf4f6fa72dcb9d8d9506145a4738ce80639b
+oid sha256:2b7c68d3a988c9b847b95c58aeb8b2e86d7ebae52088cacf0f60ffe89cf9ec7b
 size 1465

verifier-qwen3.5-2b-v2/{checkpoint-200 → checkpoint-300}/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": 0.7312614259597806,
+  "best_global_step": 250,
+  "best_metric": 0.19475513696670532,
+  "best_model_checkpoint": "./outputs/verifier-qwen3.5-2b-v2/checkpoint-250",
+  "epoch": 1.0950639853747715,
   "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": 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
+    },
+    {
+      "epoch": 0.9506398537477148,
+      "grad_norm": 1.5486634969711304,
+      "learning_rate": 5.557692307692308e-05,
+      "loss": 0.17414852380752563,
+      "step": 260
+    },
+    {
+      "epoch": 0.9872029250457038,
+      "grad_norm": 1.200329065322876,
+      "learning_rate": 5.365384615384616e-05,
+      "loss": 0.17662066221237183,
+      "step": 270
+    },
+    {
+      "epoch": 1.0219378427787935,
+      "grad_norm": 2.5651373863220215,
+      "learning_rate": 5.173076923076924e-05,
+      "loss": 0.13115617036819457,
+      "step": 280
+    },
+    {
+      "epoch": 1.0585009140767825,
+      "grad_norm": 2.1985960006713867,
+      "learning_rate": 4.980769230769231e-05,
+      "loss": 0.08628299236297607,
+      "step": 290
+    },
+    {
+      "epoch": 1.0950639853747715,
+      "grad_norm": 3.3937857151031494,
+      "learning_rate": 4.788461538461539e-05,
+      "loss": 0.10255987644195556,
+      "step": 300
+    },
+    {
+      "epoch": 1.0950639853747715,
+      "eval_loss": 0.28517603874206543,
+      "eval_runtime": 2.3859,
+      "eval_samples_per_second": 90.949,
+      "eval_steps_per_second": 11.735,
+      "step": 300
     }
   ],
   "logging_steps": 10,
@@ -199,7 +285,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6314590914852864.0,
+  "total_flos": 9490660100963328.0,
   "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:a13f30fa83f5e89c878fd30afea74be1249fb2b01cdb678a0cfeca9884336db4
-size 12008
+oid sha256:71e8b03774c377dc1a25a9706e330350fb6cd654458534d091ecaadb016421c9
+size 13334