Training in progress, step 1000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7f4907280dd14a5752e3967dc15cb696f55cac42f2bae1e4b6314a378055997a
+oid sha256:beb49b9542cd8d0592809b5da72be1065e1ed0587848ca83926f57f411f4c57f
 size 268290900

run-1/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cbd257837648e7e0d9f2a9b35ba1325ee0da4bed8cd54d9b5ecd144679e1b873
+oid sha256:beb49b9542cd8d0592809b5da72be1065e1ed0587848ca83926f57f411f4c57f
 size 268290900

run-1/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d102376c8d8c6362e33751e0a9f9653a24d89471601c10ef9b051b649ec2a2e6
+oid sha256:5c40e7e294e83f1d4d4c3ae07fa96509f55eb115aee25eb3e5c81223bd3db439
 size 536643898

run-1/checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:60ef01273aaa599804d51a93c6c0c61874ebd50ab9e135a21aab5f7cd0e6487a
+oid sha256:c794bc4c67ef18245dd516031ce405ab557e4d551d225d8dd1e1abc0f2be8e33
 size 1064

run-1/checkpoint-1000/trainer_state.json

@@ -10,50 +10,50 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5687096774193549,
-      "eval_loss": 0.1982271522283554,
-      "eval_runtime": 5.4801,
-      "eval_samples_per_second": 565.685,
-      "eval_steps_per_second": 11.861,
+      "eval_accuracy": 0.5938709677419355,
+      "eval_loss": 0.19997140765190125,
+      "eval_runtime": 5.4307,
+      "eval_samples_per_second": 570.829,
+      "eval_steps_per_second": 11.969,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5293694734573364,
-      "learning_rate": 1.4758909853249476e-05,
-      "loss": 0.3141,
+      "grad_norm": 0.5329213738441467,
+      "learning_rate": 1.685534591194969e-05,
+      "loss": 0.3229,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8093548387096774,
-      "eval_loss": 0.10087499022483826,
-      "eval_runtime": 5.6821,
-      "eval_samples_per_second": 545.571,
-      "eval_steps_per_second": 11.439,
+      "eval_accuracy": 0.8238709677419355,
+      "eval_loss": 0.09559512138366699,
+      "eval_runtime": 5.9449,
+      "eval_samples_per_second": 521.456,
+      "eval_steps_per_second": 10.934,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8661290322580645,
-      "eval_loss": 0.07097209990024567,
-      "eval_runtime": 5.487,
-      "eval_samples_per_second": 564.976,
-      "eval_steps_per_second": 11.846,
+      "eval_accuracy": 0.8812903225806452,
+      "eval_loss": 0.06311405450105667,
+      "eval_runtime": 5.4649,
+      "eval_samples_per_second": 567.253,
+      "eval_steps_per_second": 11.894,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.45407164096832275,
-      "learning_rate": 9.517819706498952e-06,
-      "loss": 0.1164,
+      "grad_norm": 0.44504690170288086,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.1111,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1908,
+  "max_steps": 3180,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 10,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -71,8 +71,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.060114748704759524,
-    "num_train_epochs": 6,
-    "temperature": 16
+    "alpha": 0.16854985287286628,
+    "num_train_epochs": 10,
+    "temperature": 10
   }
 }

run-1/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5f891ec870683b3f10d8728faaafcb50bf0331ff0eaeeefb9443043e27a8c209
+oid sha256:d3a2647be896998ae5c2f33a1c5e04b6d27594669f7060f5c0c0e4612ecd7d55
 size 5176

runs/Aug28_01-37-39_61274092231a/events.out.tfevents.1724810408.61274092231a.346.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0494892fbc7ee3d3983d53b54c79aa06b602c6839d3c1df59e5805a577b6cc35
-size 13232
+oid sha256:263907f50cce5eee49ab659218e239e2758de02fa3affed34b744da52f6fa3d3
+size 13766