Training in progress, step 475, checkpoint

last-checkpoint/adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c4096722f61b9f08b3ed290c1ad856624d55b23c6af75829f3062a4c172eadd
+oid sha256:bbf6a4de246e47839a18be854c12183b62378ed4c6156b86650a739164d8fb5a
 size 479005064

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d81bc033a5700fa27dbb81c9943b205da6e2f0eeac2c7f21a7122e7fa46074df
+oid sha256:6d9cd1120e0f3650f7abcfa9ec76e57e74a868bde3c38d6828925defc070fcb8
 size 958299770

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5134dce65814f94d568234e504c6a81c488a705fdf68f481871040907b9769ba
+oid sha256:ff25add81b6d75c22efaf6758c827de8f17e29d5144538b1fcd5701bfe6a4507
 size 14244

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31f09120e45bcc4193cfd66ac556b20e8f26a1efecd672279cdd630bf09741a4
+oid sha256:d825824fc7007a7b00c58c661fd84159061d7728ac6b0d52a8723237e0ac3962
 size 1064

last-checkpoint/trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.0035839153877398238,
+  "epoch": 0.003783021798169814,
   "eval_steps": 500,
-  "global_step": 450,
+  "global_step": 475,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -547,6 +547,30 @@
       "reward_std": 0.17024155631661414,
       "rewards/custom_reward_logic_v4_batch_streak": 0.10238021239638329,
       "step": 450
+    },
+    {
+      "completion_length": 39.55,
+      "epoch": 0.0036635579519118198,
+      "grad_norm": 0.2565569579601288,
+      "kl": 0.33122892007231713,
+      "learning_rate": 4.873106608932585e-06,
+      "loss": 0.0132,
+      "reward": 0.1605322863906622,
+      "reward_std": 0.3534850224852562,
+      "rewards/custom_reward_logic_v4_batch_streak": 0.1605322863906622,
+      "step": 460
+    },
+    {
+      "completion_length": 20.33125,
+      "epoch": 0.003743200516083816,
+      "grad_norm": 0.9290266633033752,
+      "kl": 0.3362825021147728,
+      "learning_rate": 4.861418843169012e-06,
+      "loss": 0.0135,
+      "reward": 0.06937500275671482,
+      "reward_std": 0.1000722162425518,
+      "rewards/custom_reward_logic_v4_batch_streak": 0.06937500275671482,
+      "step": 470
     }
   ],
   "logging_steps": 10,