Training in progress, step 500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a22c8dadae3169476a58e86197c64b4855ed3edd99cb31a9fe32e1fdb79687a8
+oid sha256:cc8c088dd2439cacfe1e095ae943fafbed28ec1cabf8664b50579745419bf06e
 size 598898116

run-1/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:749188f365e911f11256fe2945c70c092d551fab90fc399eb8b9c37c89046ee0
+oid sha256:cc8c088dd2439cacfe1e095ae943fafbed28ec1cabf8664b50579745419bf06e
 size 598898116

run-1/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a71d877b7f777422c9898fd486ca34976de9cf71a59fb99937afd46c8b426cce
+oid sha256:67a9bc1cb444595afd4387a247d83b5259d39585d2e460af5046ec910cdef4eb
 size 1197884026

run-1/checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:306c6af96669ed13365a61a72a74a28f181e5e20ceb2078a041489ebbb847646
+oid sha256:d73e26459c92e195a7738eb8b675e4359a54bd79fe9532dd9d1f04483172c359
 size 1064

run-1/checkpoint-500/trainer_state.json

@@ -11,50 +11,50 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.7729032258064517,
-      "eval_loss": 3.520946979522705,
-      "eval_runtime": 20.0431,
-      "eval_samples_per_second": 154.667,
-      "eval_steps_per_second": 1.646,
+      "eval_accuracy": 0.6248387096774194,
+      "eval_loss": 2.5749671459198,
+      "eval_runtime": 22.9474,
+      "eval_samples_per_second": 135.092,
+      "eval_steps_per_second": 1.438,
       "step": 159
     },
     {
       "epoch": 1.2578616352201257,
-      "grad_norm": 6.797354698181152,
-      "learning_rate": 1.5828092243186584e-05,
-      "loss": 5.8294,
+      "grad_norm": 5.373847961425781,
+      "learning_rate": 1.49937106918239e-05,
+      "loss": 3.5751,
       "step": 200
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.9245161290322581,
-      "eval_loss": 1.7692639827728271,
-      "eval_runtime": 20.2508,
-      "eval_samples_per_second": 153.081,
-      "eval_steps_per_second": 1.63,
+      "eval_accuracy": 0.8938709677419355,
+      "eval_loss": 1.5459049940109253,
+      "eval_runtime": 22.8738,
+      "eval_samples_per_second": 135.526,
+      "eval_steps_per_second": 1.443,
       "step": 318
     },
     {
       "epoch": 2.5157232704402515,
-      "grad_norm": 4.905786514282227,
-      "learning_rate": 1.1635220125786164e-05,
-      "loss": 1.9917,
+      "grad_norm": 3.5598983764648438,
+      "learning_rate": 9.962264150943397e-06,
+      "loss": 1.6838,
       "step": 400
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.9438709677419355,
-      "eval_loss": 1.2671879529953003,
-      "eval_runtime": 20.1246,
-      "eval_samples_per_second": 154.04,
-      "eval_steps_per_second": 1.64,
+      "eval_accuracy": 0.9290322580645162,
+      "eval_loss": 1.163897156715393,
+      "eval_runtime": 22.8793,
+      "eval_samples_per_second": 135.494,
+      "eval_steps_per_second": 1.442,
       "step": 477
     }
   ],
   "logging_steps": 200,
-  "max_steps": 954,
+  "max_steps": 795,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 6,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -72,8 +72,8 @@
   "train_batch_size": 96,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.4005807540566969,
-    "num_train_epochs": 6,
-    "temperature": 8
+    "alpha": 0.30165501089690516,
+    "num_train_epochs": 5,
+    "temperature": 19
   }
 }

run-1/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:166ac319de89d357910c885138ea139bcf526da11716bd9af418e8fabca59416
+oid sha256:0f2a6043dcb359bf6e5c547e39efa179f167939f5b2be51fd4993780a524ca74
 size 5368

runs/Sep28_10-21-38_bb2384aee55a/events.out.tfevents.1759054927.bb2384aee55a.76.1

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:57ec3628ce3320a9ec6e249667d33856d640f0486e9f42d916989fc94ca5f597
-size 29687
+oid sha256:917c5feb177169e9871fa7d66bc2801741e1567f4fe7880724d0d0841478b737
+size 30364

runs/Sep28_10-21-38_bb2384aee55a/events.out.tfevents.1759056641.bb2384aee55a.76.2

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7adeae347ab7f5269da99be778fbbefeb0917578f24213cebcfb1ddd2f4e64fc
+size 14501

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1b5ead62771fb4a133ea7812d8458dd760625e18862235c9bfe1d87fb1d371e
+oid sha256:0f2a6043dcb359bf6e5c547e39efa179f167939f5b2be51fd4993780a524ca74
 size 5368