Training in progress, step 1500

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:beb49b9542cd8d0592809b5da72be1065e1ed0587848ca83926f57f411f4c57f
+oid sha256:c56142fe1f54690c5d2f361924ff76d63ced6daf99acb3e6fd3eee9cda2b7c33
 size 268290900

run-1/checkpoint-1500/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b83f5982945de1d49d0a4cf8a401b11dd21131d57b38e37bb85e2b887d34d2c5
+oid sha256:c56142fe1f54690c5d2f361924ff76d63ced6daf99acb3e6fd3eee9cda2b7c33
 size 268290900

run-1/checkpoint-1500/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:35f1f3775a7446f821bfb6b80a7c2af540e9f90d5b4b2e964e6434b31ed3c9ef
+oid sha256:f4421fbe9438d2b39e07121b2a343e31519636078fcaa35b8ce93f3a6130e08f
 size 536643898

run-1/checkpoint-1500/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:55c8d3ce0734337fc0c187ca5543b4c70ca45d996531f199209b3a0c2a798109
+oid sha256:71453465aad25f4c5a0a948496c64b1f74df850abda497954afe3695c00756ee
 size 1064

run-1/checkpoint-1500/trainer_state.json

@@ -10,66 +10,66 @@
   "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
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.89,
-      "eval_loss": 0.05795975774526596,
-      "eval_runtime": 5.584,
-      "eval_samples_per_second": 555.161,
-      "eval_steps_per_second": 11.64,
+      "eval_accuracy": 0.9006451612903226,
+      "eval_loss": 0.048286207020282745,
+      "eval_runtime": 5.6179,
+      "eval_samples_per_second": 551.809,
+      "eval_steps_per_second": 11.57,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.3293832242488861,
-      "learning_rate": 4.276729559748428e-06,
-      "loss": 0.0802,
+      "grad_norm": 0.3021944761276245,
+      "learning_rate": 1.0566037735849058e-05,
+      "loss": 0.0699,
       "step": 1500
     }
   ],
   "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": {
@@ -87,8 +87,8 @@
   "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-1500/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 sha256:263907f50cce5eee49ab659218e239e2758de02fa3affed34b744da52f6fa3d3
-size 13766
+oid sha256:69d08b21476c340fdf5698ce4c3a4f98a8a076dc99b6f52dc6fc0fc87876aec4
+size 14623