Training in progress, step 1000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8c85b42126896ca0a309e576edad67efd01ee7456c9094190b7090f8d41a2ef7
+oid sha256:eeda0e892612539f228a2d85b9ed5095ecc8961d9c39074dec9e8086ce6ea567
 size 268290900

run-1/checkpoint-1000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6865e052f59a33ee2d3498562ba49d6ab9c2e8aee5a322f1ac4e927a15ba4612
+oid sha256:eeda0e892612539f228a2d85b9ed5095ecc8961d9c39074dec9e8086ce6ea567
 size 268290900

run-1/checkpoint-1000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e9e6d6e069262a0e9683b7631969201809882fc02f9caf4149ef1ce8e7e8059
+oid sha256:2333b1b2fe41da59806f818c74eb9eae5be456ef45dad7e5846da0a836b36a63
 size 536643898

run-1/checkpoint-1000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9f182501c34e4ea3ebc7617d27edab7e1367582b147e518cd90295ec7f2eaa0f
+oid sha256:045cf4abe065257226fe11ac1a91dfd4c5b953cb2366365ef8bb18b4cf33e734
 size 1064

run-1/checkpoint-1000/trainer_state.json

@@ -12,15 +12,15 @@
       "epoch": 1.0,
       "eval_accuracy": 0.0064516129032258064,
       "eval_loss": NaN,
-      "eval_runtime": 5.3104,
-      "eval_samples_per_second": 583.757,
-      "eval_steps_per_second": 12.24,
+      "eval_runtime": 5.354,
+      "eval_samples_per_second": 579.009,
+      "eval_steps_per_second": 12.141,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 94451.3046875,
-      "learning_rate": 1.371069182389937e-05,
+      "grad_norm": 143768.90625,
+      "learning_rate": 1.606918238993711e-05,
       "loss": 0.0,
       "step": 500
     },
@@ -28,32 +28,32 @@
       "epoch": 2.0,
       "eval_accuracy": 0.0064516129032258064,
       "eval_loss": NaN,
-      "eval_runtime": 5.4472,
-      "eval_samples_per_second": 569.1,
-      "eval_steps_per_second": 11.933,
+      "eval_runtime": 5.464,
+      "eval_samples_per_second": 567.352,
+      "eval_steps_per_second": 11.896,
       "step": 636
     },
     {
       "epoch": 3.0,
       "eval_accuracy": 0.0064516129032258064,
       "eval_loss": NaN,
-      "eval_runtime": 5.4183,
-      "eval_samples_per_second": 572.14,
-      "eval_steps_per_second": 11.996,
+      "eval_runtime": 5.2458,
+      "eval_samples_per_second": 590.953,
+      "eval_steps_per_second": 12.391,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 233628.765625,
-      "learning_rate": 7.421383647798742e-06,
+      "grad_norm": 371973.03125,
+      "learning_rate": 1.2138364779874214e-05,
       "loss": 0.0,
       "step": 1000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 1590,
+  "max_steps": 2544,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 8,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -71,8 +71,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.16176343022645268,
-    "num_train_epochs": 5,
-    "temperature": 12
+    "alpha": 0.41600283894248313,
+    "num_train_epochs": 8,
+    "temperature": 20
   }
 }

run-1/checkpoint-1000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:294e8e48bc1549edad3189588f706dce11934ea0b8d7c455022c6664ea0826eb
+oid sha256:e6722fa4d4cd319ee6d313948dfb65a13aa608029a7cedd96bdd0a50d4d1c279
 size 5304

runs/Nov11_11-16-16_921d9136ef3a/events.out.tfevents.1731325170.921d9136ef3a.798.5

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b33900cfb728310283b0f933b19d8d43fe2fd6a6518b00893ce90604c78d6c4d
-size 13277
+oid sha256:8de1709c887b853ebe84890b2ef2cccc34ea75ba495f1e5ceae5e8690f953e88
+size 14134