checkpoint-200

verifier-qwen3.5-2b-v2/{checkpoint-100 → checkpoint-200}/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e6d2ba7395cdf406bc8be4aa21f2fd2e2adab2bcad7ba43361f29aef0e5920a2
+oid sha256:5f64d0d61ae7227cd0a4930a430d1c31fd05a3437c5d4a263e04106d2ac27179
 size 134604152

verifier-qwen3.5-2b-v2/{checkpoint-100 → checkpoint-200}/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f00de01a64a0c6e343865937151737caa4447caa09458b5c49157ac6c2973dc2
+oid sha256:2798f5617a6babee74f7dd725c87d5c32580bb157447fb4b7e070022175ac321
 size 269426431

verifier-qwen3.5-2b-v2/{checkpoint-100 → checkpoint-200}/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fe87b8c009bf7995c9e86c4e59f1cf48c6750944b0ccdacf4db971c9bd16e7e1
+oid sha256:214c39d274aae7a84abefce418c08173a922fa2de0e1a2bad24048cdbd4c2692
 size 14645

verifier-qwen3.5-2b-v2/{checkpoint-100 → checkpoint-200}/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e85762b23b88c4fa770db2d706ff9af49a9222a1a72d1536da2849fd5f72bf53
+oid sha256:9345529b6e002b72948f483d9aa5bf4f6fa72dcb9d8d9506145a4738ce80639b
 size 1465

verifier-qwen3.5-2b-v2/{checkpoint-100 → checkpoint-200}/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 100,
-  "best_metric": 0.20814214646816254,
-  "best_model_checkpoint": "./outputs/verifier-qwen3.5-2b-v2/checkpoint-100",
-  "epoch": 0.3656307129798903,
+  "best_global_step": 150,
+  "best_metric": 0.19720077514648438,
+  "best_model_checkpoint": "./outputs/verifier-qwen3.5-2b-v2/checkpoint-150",
+  "epoch": 0.7312614259597806,
   "eval_steps": 50,
-  "global_step": 100,
+  "global_step": 200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -94,6 +94,92 @@
       "eval_samples_per_second": 91.174,
       "eval_steps_per_second": 11.764,
       "step": 100
+    },
+    {
+      "epoch": 0.40219378427787933,
+      "grad_norm": 3.3533239364624023,
+      "learning_rate": 8.442307692307693e-05,
+      "loss": 0.1813126564025879,
+      "step": 110
+    },
+    {
+      "epoch": 0.43875685557586835,
+      "grad_norm": 1.6487245559692383,
+      "learning_rate": 8.25e-05,
+      "loss": 0.20677659511566163,
+      "step": 120
+    },
+    {
+      "epoch": 0.4753199268738574,
+      "grad_norm": 2.074784755706787,
+      "learning_rate": 8.057692307692308e-05,
+      "loss": 0.20372426509857178,
+      "step": 130
+    },
+    {
+      "epoch": 0.5118829981718465,
+      "grad_norm": 2.7923927307128906,
+      "learning_rate": 7.865384615384616e-05,
+      "loss": 0.18218755722045898,
+      "step": 140
+    },
+    {
+      "epoch": 0.5484460694698354,
+      "grad_norm": 1.6166132688522339,
+      "learning_rate": 7.673076923076924e-05,
+      "loss": 0.217415714263916,
+      "step": 150
+    },
+    {
+      "epoch": 0.5484460694698354,
+      "eval_loss": 0.19720077514648438,
+      "eval_runtime": 2.3834,
+      "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
     }
   ],
   "logging_steps": 10,
@@ -113,7 +199,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3160604495935488.0,
+  "total_flos": 6314590914852864.0,
   "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:b255f51f1b42be939e9ca659a3c51c47953fe9b5f99090894efabe26feda02ce
-size 9356
+oid sha256:d799b7e149ac948bf7e29e98cc328c71a480d5ecbcf9b6c77b5f59d823622f57
+size 10682