Training in progress, step 375, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:eb02bc2606b7be732030267e54281bba87f367c8724f47940de822f8002ab5a7
+oid sha256:504a947a76d4e9379f592d23ac1288d770e5938977609fe24ec6c607d11046ce
 size 479005064

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:648578a4fd856f7cd2067223584de606b38116d27f3e955649a638c2114f2fa4
+oid sha256:44e9eb6cf37611d240b4204985eb5221aaf601e43cb08745edaa939eb10d6c65
 size 958299770

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9247630071748e54bed953c954732cd9f6db56835e2b1e59df44567e0764e2c3
+oid sha256:27123da8aeb5b353cabf5cb39e63885135b4041099e95c98394b966cce5871e7
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8b017f223f7cbfcbd91ec9cfbd378a9bc9c27e3fa5227c019815cc49a05577b2
+oid sha256:38ee62f39915fab64d2f1ab670d54e213cfc2d70e4cd48a7b1400797011d55aa
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.002787489746019863,
+  "epoch": 0.002986596156449853,
   "eval_steps": 500,
-  "global_step": 350,
+  "global_step": 375,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -427,6 +427,30 @@
       "reward_std": 0.10931163281202316,
       "rewards/custom_reward_logic_v4_batch_streak": 0.2172499977052212,
       "step": 350
+    },
+    {
+      "completion_length": 18.9375,
+      "epoch": 0.002867132310191859,
+      "grad_norm": 0.5375373363494873,
+      "kl": 0.33093191757798196,
+      "learning_rate": 4.962019382530521e-06,
+      "loss": 0.0132,
+      "reward": 0.3819999981671572,
+      "reward_std": 0.1661699414253235,
+      "rewards/custom_reward_logic_v4_batch_streak": 0.3819999981671572,
+      "step": 360
+    },
+    {
+      "completion_length": 18.0,
+      "epoch": 0.002946774874363855,
+      "grad_norm": 1.0857776403427124,
+      "kl": 0.3886502429842949,
+      "learning_rate": 4.955445176176577e-06,
+      "loss": 0.0155,
+      "reward": 0.17512499764561654,
+      "reward_std": 0.16201305240392685,
+      "rewards/custom_reward_logic_v4_batch_streak": 0.17512499764561654,
+      "step": 370
     }
   ],
   "logging_steps": 10,