Training in progress, step 4000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6230f90c11cc2d67209cb5e4ad54d202bbb6434781258269e6a925fa5377e00d
+oid sha256:ae7335096cf5b1fd84c6e82b5759ed57ddf44a0e121f4e3c590a965600f0f28e
 size 498612824

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9878449e793507c763b443b91a6a37e8d101ae2a4c9b8b59601b054e59a97709
+oid sha256:0b17f5e96de107e606342fe64d95fbbb83b976cdfb9761f3be3f5ff3925bb785
 size 997345931

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:abf2b0fc64c4be716a75868bdb69c6ff3620ba2a21085b2c861288da2529d5e7
+oid sha256:dd8861c76c95a138fb74b2279a575c494edce4804040a54509842ed6fd8b90fe
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5ac1c46a2776d12775d23d0f587efc112188137ce2140da35bc15d301c9f620e
+oid sha256:c0033c7745b46bdca3ecab5787678834ca68f7f7e1288869dceeb38812abc253
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7fe4d93dfc6f647f747e92a74d9d5986e59893a476d774e2205030dc67d6ff6d
+oid sha256:012264319a2f0e83b703857fe0c90509f0ba126782a5ba9dccbac93f5b664216
 size 1465

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 3000,
-  "best_metric": 0.12173645943403244,
-  "best_model_checkpoint": "./training_output/checkpoint-3000",
-  "epoch": 0.15,
+  "best_global_step": 4000,
+  "best_metric": 0.04343162104487419,
+  "best_model_checkpoint": "./training_output/checkpoint-4000",
+  "epoch": 0.2,
   "eval_steps": 1000,
-  "global_step": 3000,
+  "global_step": 4000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -77,6 +77,29 @@
       "eval_samples_per_second": 55.123,
       "eval_steps_per_second": 3.445,
       "step": 3000
+    },
+    {
+      "epoch": 0.175,
+      "grad_norm": 37.91666793823242,
+      "learning_rate": 1.6501e-05,
+      "loss": 0.0647,
+      "step": 3500
+    },
+    {
+      "epoch": 0.2,
+      "grad_norm": 0.18077421188354492,
+      "learning_rate": 1.6001e-05,
+      "loss": 0.0587,
+      "step": 4000
+    },
+    {
+      "epoch": 0.2,
+      "eval_accuracy": 0.9575,
+      "eval_loss": 0.04343162104487419,
+      "eval_runtime": 361.6433,
+      "eval_samples_per_second": 55.303,
+      "eval_steps_per_second": 3.456,
+      "step": 4000
     }
   ],
   "logging_steps": 500,
@@ -96,7 +119,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 3157332664320000.0,
+  "total_flos": 4209776885760000.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null