Training in progress, step 1500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0a5e1d970acbb5d044c7b95b544b2dbcf4680367175c4e62724a8bf3391cbe35
+oid sha256:9d4f707e65d41640c3ac73464d02cb5f11e66ddf6de0e369ed9a9a5d57c45bc7
 size 268290900

run-4/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2ee948a20c89755d7a96becd052e4426cb6cdccf9b8ce076c95e8904c6697ce5
+oid sha256:85f37b02e3883320516648328ce14dafe511cb4589da34cfa07e2dfa2476622a
 size 268290900

run-4/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ceae458e1139240eed9e7d904d0cad7cb3b3fe26a87e4014a66ee16b5ed87def
+oid sha256:fa59e9b6c2b95d771d0ce88b537db679baa097351c691e48e23e8387f2a0ae28
 size 536643898

run-4/checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ed9da1cbe846856af516cf037dd225e7f48e2737ef84c42a9bee0753b8c140ef
+oid sha256:045cf4abe065257226fe11ac1a91dfd4c5b953cb2366365ef8bb18b4cf33e734
 size 1064

run-4/checkpoint-1000/trainer_state.json

@@ -10,50 +10,50 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6445161290322581,
-      "eval_loss": 0.29249975085258484,
-      "eval_runtime": 5.1547,
-      "eval_samples_per_second": 601.395,
-      "eval_steps_per_second": 12.61,
+      "eval_accuracy": 0.6464516129032258,
+      "eval_loss": 0.2894870638847351,
+      "eval_runtime": 5.1625,
+      "eval_samples_per_second": 600.482,
+      "eval_steps_per_second": 12.591,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.6633772850036621,
-      "learning_rate": 1.550763701707098e-05,
-      "loss": 0.4626,
+      "grad_norm": 0.6621634364128113,
+      "learning_rate": 1.606918238993711e-05,
+      "loss": 0.4605,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8380645161290322,
-      "eval_loss": 0.12248263508081436,
-      "eval_runtime": 5.377,
-      "eval_samples_per_second": 576.528,
-      "eval_steps_per_second": 12.088,
+      "eval_accuracy": 0.8416129032258064,
+      "eval_loss": 0.11869335174560547,
+      "eval_runtime": 5.2214,
+      "eval_samples_per_second": 593.71,
+      "eval_steps_per_second": 12.449,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8887096774193548,
-      "eval_loss": 0.07343784719705582,
-      "eval_runtime": 5.2352,
-      "eval_samples_per_second": 592.149,
-      "eval_steps_per_second": 12.416,
+      "eval_accuracy": 0.8912903225806451,
+      "eval_loss": 0.0700218454003334,
+      "eval_runtime": 5.1185,
+      "eval_samples_per_second": 605.652,
+      "eval_steps_per_second": 12.699,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.59869384765625,
-      "learning_rate": 1.101527403414196e-05,
-      "loss": 0.1438,
+      "grad_norm": 0.5828885436058044,
+      "learning_rate": 1.2138364779874214e-05,
+      "loss": 0.1397,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2226,
+  "max_steps": 2544,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 8,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -71,8 +71,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.3308965359344769,
-    "num_train_epochs": 7,
+    "alpha": 0.6565754742921168,
+    "num_train_epochs": 8,
     "temperature": 3
   }
 }

run-4/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b44c3bc0d3a8cfab92f5679709a7e42f1cbbf073f24e273be0df24113c15c6e9
+oid sha256:954f663a36c0409373a26eb89ffe90cd8c95b2e47593d56f176a8af75ce45afe
 size 5240

run-4/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c9a63d3004610f74819b6489ca4b580024510d89b7576222160d376a94edd03
+oid sha256:9d4f707e65d41640c3ac73464d02cb5f11e66ddf6de0e369ed9a9a5d57c45bc7
 size 268290900

run-4/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:90b0a266aa6d2c7cee70af38c41264e284f7c2475c5c6c36d6483a6cc0cab912
+oid sha256:c57ee7395fd8395ea43278ecb5a3a37fb0b6f39213bee3b1f81726d248d1c4fb
 size 536643898

run-4/checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ad4d7d251acf36e559c362893a1fb310c9f46b20e8a330025a14b6829ce4ab07
+oid sha256:eb7b4645e59a5c5cb66f63e8bcb5f006c535fd0ea63db9ae152dd586fd465b28
 size 1064

run-4/checkpoint-1500/trainer_state.json

@@ -10,66 +10,66 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.6445161290322581,
-      "eval_loss": 0.29249975085258484,
-      "eval_runtime": 5.1547,
-      "eval_samples_per_second": 601.395,
-      "eval_steps_per_second": 12.61,
+      "eval_accuracy": 0.6464516129032258,
+      "eval_loss": 0.2894870638847351,
+      "eval_runtime": 5.1625,
+      "eval_samples_per_second": 600.482,
+      "eval_steps_per_second": 12.591,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.6633772850036621,
-      "learning_rate": 1.550763701707098e-05,
-      "loss": 0.4626,
+      "grad_norm": 0.6621634364128113,
+      "learning_rate": 1.606918238993711e-05,
+      "loss": 0.4605,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.8380645161290322,
-      "eval_loss": 0.12248263508081436,
-      "eval_runtime": 5.377,
-      "eval_samples_per_second": 576.528,
-      "eval_steps_per_second": 12.088,
+      "eval_accuracy": 0.8416129032258064,
+      "eval_loss": 0.11869335174560547,
+      "eval_runtime": 5.2214,
+      "eval_samples_per_second": 593.71,
+      "eval_steps_per_second": 12.449,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8887096774193548,
-      "eval_loss": 0.07343784719705582,
-      "eval_runtime": 5.2352,
-      "eval_samples_per_second": 592.149,
-      "eval_steps_per_second": 12.416,
+      "eval_accuracy": 0.8912903225806451,
+      "eval_loss": 0.0700218454003334,
+      "eval_runtime": 5.1185,
+      "eval_samples_per_second": 605.652,
+      "eval_steps_per_second": 12.699,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.59869384765625,
-      "learning_rate": 1.101527403414196e-05,
-      "loss": 0.1438,
+      "grad_norm": 0.5828885436058044,
+      "learning_rate": 1.2138364779874214e-05,
+      "loss": 0.1397,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9045161290322581,
-      "eval_loss": 0.05444410815834999,
-      "eval_runtime": 5.4434,
-      "eval_samples_per_second": 569.5,
-      "eval_steps_per_second": 11.941,
+      "eval_accuracy": 0.9077419354838709,
+      "eval_loss": 0.05109017714858055,
+      "eval_runtime": 5.3994,
+      "eval_samples_per_second": 574.138,
+      "eval_steps_per_second": 12.038,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.4075869023799896,
-      "learning_rate": 6.522911051212939e-06,
-      "loss": 0.0845,
+      "grad_norm": 0.38680753111839294,
+      "learning_rate": 8.207547169811321e-06,
+      "loss": 0.0802,
       "step": 1500
     }
   ],
   "logging_steps": 500,
-  "max_steps": 2226,
+  "max_steps": 2544,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 7,
+  "num_train_epochs": 8,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -87,8 +87,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.3308965359344769,
-    "num_train_epochs": 7,
+    "alpha": 0.6565754742921168,
+    "num_train_epochs": 8,
     "temperature": 3
   }
 }

run-4/checkpoint-1500/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b44c3bc0d3a8cfab92f5679709a7e42f1cbbf073f24e273be0df24113c15c6e9
+oid sha256:954f663a36c0409373a26eb89ffe90cd8c95b2e47593d56f176a8af75ce45afe
 size 5240

runs/Oct11_17-03-17_c179252869f4/events.out.tfevents.1728668933.c179252869f4.732.5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7e3c91fbf5869ac9b7e6db6fd2f767702b53f072436346b3fe121a294ec71579
-size 14155
+oid sha256:5546059f8ea9ff7fa0c8dca592c8b8f6f87ef6ffef099ef0d0f5127f7e757c29
+size 15546