Training in progress, step 3000, checkpoint

last-checkpoint/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:adadf44f1e459344e83749883329a7a42d9cd526e045515d581288a614c95052
+oid sha256:6230f90c11cc2d67209cb5e4ad54d202bbb6434781258269e6a925fa5377e00d
 size 498612824

last-checkpoint/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ca0939f144f4a3fab18378090a7055b038e279c6b9e2ff8a9a2a8ff2a99c8b8c
+oid sha256:9878449e793507c763b443b91a6a37e8d101ae2a4c9b8b59601b054e59a97709
 size 997345931

last-checkpoint/rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8cbb9d20ae0e060525bcac1b70551f336242a3392c3ba0392910b81075e6dc33
+oid sha256:abf2b0fc64c4be716a75868bdb69c6ff3620ba2a21085b2c861288da2529d5e7
 size 14645

last-checkpoint/scaler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f4aa03f6e0cd07cf67ce1fbe3101d545f5771ef9148b9debf02b11cf6948da5c
+oid sha256:5ac1c46a2776d12775d23d0f587efc112188137ce2140da35bc15d301c9f620e
 size 1383

last-checkpoint/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f8cac132f6cb7e26c16168d7dd2d5889b5966954ea1f2f629ae0cf0022b48d02
+oid sha256:7fe4d93dfc6f647f747e92a74d9d5986e59893a476d774e2205030dc67d6ff6d
 size 1465

last-checkpoint/trainer_state.json

@@ -1,10 +1,10 @@
 {
-  "best_global_step": 1000,
-  "best_metric": 0.12729284167289734,
-  "best_model_checkpoint": "./training_output/checkpoint-1000",
-  "epoch": 0.1,
+  "best_global_step": 3000,
+  "best_metric": 0.12173645943403244,
+  "best_model_checkpoint": "./training_output/checkpoint-3000",
+  "epoch": 0.15,
   "eval_steps": 1000,
-  "global_step": 2000,
+  "global_step": 3000,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -54,6 +54,29 @@
       "eval_samples_per_second": 55.087,
       "eval_steps_per_second": 3.443,
       "step": 2000
+    },
+    {
+      "epoch": 0.125,
+      "grad_norm": 0.02591518871486187,
+      "learning_rate": 1.7501e-05,
+      "loss": 0.0668,
+      "step": 2500
+    },
+    {
+      "epoch": 0.15,
+      "grad_norm": 0.07041072845458984,
+      "learning_rate": 1.7001000000000002e-05,
+      "loss": 0.0606,
+      "step": 3000
+    },
+    {
+      "epoch": 0.15,
+      "eval_accuracy": 0.9291,
+      "eval_loss": 0.12173645943403244,
+      "eval_runtime": 362.8269,
+      "eval_samples_per_second": 55.123,
+      "eval_steps_per_second": 3.445,
+      "step": 3000
     }
   ],
   "logging_steps": 500,
@@ -73,7 +96,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 2104888442880000.0,
+  "total_flos": 3157332664320000.0,
   "train_batch_size": 2,
   "trial_name": null,
   "trial_params": null