Training in progress, step 500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d2b79ce96a29dfd90b469227636d4c69f39baa91d8c602442edf0d95debb71c6
+oid sha256:294322a3921efabcf477dc40b7b403c369a45a4957c3e80807fc97e64c4a09cf
 size 268290900

run-2/checkpoint-1000/trainer_state.json

@@ -10,50 +10,50 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.587741935483871,
-      "eval_loss": 0.19460442662239075,
-      "eval_runtime": 5.4234,
-      "eval_samples_per_second": 571.599,
-      "eval_steps_per_second": 11.985,
+      "eval_accuracy": 0.6690322580645162,
+      "eval_loss": 0.42948082089424133,
+      "eval_runtime": 5.3921,
+      "eval_samples_per_second": 574.917,
+      "eval_steps_per_second": 12.055,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5228433012962341,
-      "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3144,
+      "grad_norm": 0.8646999001502991,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.6591,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8212903225806452,
-      "eval_loss": 0.09393113851547241,
-      "eval_runtime": 5.9301,
-      "eval_samples_per_second": 522.759,
-      "eval_steps_per_second": 10.961,
+      "eval_accuracy": 0.8229032258064516,
+      "eval_loss": 0.16900987923145294,
+      "eval_runtime": 5.4801,
+      "eval_samples_per_second": 565.687,
+      "eval_steps_per_second": 11.861,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8793548387096775,
-      "eval_loss": 0.06259110569953918,
-      "eval_runtime": 5.99,
-      "eval_samples_per_second": 517.529,
-      "eval_steps_per_second": 10.851,
+      "eval_accuracy": 0.8729032258064516,
+      "eval_loss": 0.0963619202375412,
+      "eval_runtime": 5.6559,
+      "eval_samples_per_second": 548.098,
+      "eval_steps_per_second": 11.492,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4349106550216675,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.109,
+      "grad_norm": 0.7748064398765564,
+      "learning_rate": 7.421383647798742e-06,
+      "loss": 0.1958,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3180,
+  "max_steps": 1590,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -71,8 +71,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.11018489395767606,
-    "num_train_epochs": 10,
-    "temperature": 13
+    "alpha": 0.1651888586206014,
+    "num_train_epochs": 5,
+    "temperature": 2
   }
 }

run-2/checkpoint-500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:cd23b96be7e9aeadde19459e2d9f65b515679421480dfd74d9a3d3c96f50456d
+oid sha256:294322a3921efabcf477dc40b7b403c369a45a4957c3e80807fc97e64c4a09cf
 size 268290900

run-2/checkpoint-500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:968bf33e52d67e7f5bc0284c9b15119d0d36792cf76bafe4b7c7dc6d084a9bac
+oid sha256:5c90eb90b8269b7374850d7f98358b4fcfca644626c4d7e9cece862afe8f3803
 size 536643898

run-2/checkpoint-500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:04366f62f8f88f5a8265df59adb051b320463277845db80e7fa43f13110c18c9
+oid sha256:9e1264523e958cf7990dc5f42d876cc12129475c4603804cf66868aaf25c2c24
 size 1064

run-2/checkpoint-500/trainer_state.json

@@ -10,25 +10,25 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.587741935483871,
-      "eval_loss": 0.19460442662239075,
-      "eval_runtime": 5.4234,
-      "eval_samples_per_second": 571.599,
-      "eval_steps_per_second": 11.985,
+      "eval_accuracy": 0.6690322580645162,
+      "eval_loss": 0.42948082089424133,
+      "eval_runtime": 5.3921,
+      "eval_samples_per_second": 574.917,
+      "eval_steps_per_second": 12.055,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5228433012962341,
-      "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3144,
+      "grad_norm": 0.8646999001502991,
+      "learning_rate": 1.371069182389937e-05,
+      "loss": 0.6591,
       "step": 500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3180,
+  "max_steps": 1590,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 5,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -46,8 +46,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.11018489395767606,
-    "num_train_epochs": 10,
-    "temperature": 13
+    "alpha": 0.1651888586206014,
+    "num_train_epochs": 5,
+    "temperature": 2
   }
 }

run-2/checkpoint-500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54a21424fcc7b0ab6ff9f908cdab480c20a3c07931af6e79061f4c4e5b5311db
+oid sha256:e3bb46aada442b39115d169cd9c125713b5e861f4eaf782c3cede72bc75c5fe3
 size 5304

runs/Nov12_03-50-03_fde6a76996dd/events.out.tfevents.1731385960.fde6a76996dd.16977.4

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7299387b0655afa7853258c09274f2fb1a3e6607c6a114e90ca322d5189eb5a7
+size 13811

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1a2c86f2e5c80ae7557f0774f6f26d6131e8b67e4418482403db97fa0118b688
+oid sha256:e3bb46aada442b39115d169cd9c125713b5e861f4eaf782c3cede72bc75c5fe3
 size 5304