Training in progress, step 2000

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f9b90474c183b5df91636679be91a2c4bf97301a8ca21ab1cefdaff8b2a19677
+oid sha256:a217643cc377425597661e1208d8138b8b1aee59dd796065b08c20e9b6c924eb
 size 268290900

run-1/checkpoint-2000/model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0d3a72e6f3052a969150c15331855655e3cc37adb5c7b717c294c90779ee2451
+oid sha256:a217643cc377425597661e1208d8138b8b1aee59dd796065b08c20e9b6c924eb
 size 268290900

run-1/checkpoint-2000/optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:270ebab16da48210033fb5c268c0c452cefef9bec7254070e1ce9c77bbac3b96
+oid sha256:2f8d27f47c720dc2f46048728fce7744fd99c57566e55b55570564aa3d2a4397
 size 536643898

run-1/checkpoint-2000/scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43d45918bfeb622ba899798234ec7646c7a90dd3f5771db086f5b7dee1d5a530
+oid sha256:2bf6d4c2c13eda2c2dfacdbfc8684055c5603a92b6c0c386d7c757b108195d12
 size 1064

run-1/checkpoint-2000/trainer_state.json

@@ -10,91 +10,91 @@
   "log_history": [
     {
       "epoch": 1.0,
-      "eval_accuracy": 0.5838709677419355,
-      "eval_loss": 0.19346614181995392,
-      "eval_runtime": 5.1609,
-      "eval_samples_per_second": 600.675,
-      "eval_steps_per_second": 12.595,
+      "eval_accuracy": 0.5829032258064516,
+      "eval_loss": 0.1985393762588501,
+      "eval_runtime": 5.6203,
+      "eval_samples_per_second": 551.572,
+      "eval_steps_per_second": 11.565,
       "step": 318
     },
     {
       "epoch": 1.5723270440251573,
-      "grad_norm": 0.5179193019866943,
-      "learning_rate": 1.685534591194969e-05,
-      "loss": 0.3125,
+      "grad_norm": 0.5288612246513367,
+      "learning_rate": 1.606918238993711e-05,
+      "loss": 0.3184,
       "step": 500
     },
     {
       "epoch": 2.0,
-      "eval_accuracy": 0.82,
-      "eval_loss": 0.09356740862131119,
-      "eval_runtime": 5.2878,
-      "eval_samples_per_second": 586.255,
-      "eval_steps_per_second": 12.292,
+      "eval_accuracy": 0.8174193548387096,
+      "eval_loss": 0.09726663678884506,
+      "eval_runtime": 6.0438,
+      "eval_samples_per_second": 512.92,
+      "eval_steps_per_second": 10.755,
       "step": 636
     },
     {
       "epoch": 3.0,
-      "eval_accuracy": 0.8790322580645161,
-      "eval_loss": 0.06245172396302223,
-      "eval_runtime": 5.0672,
-      "eval_samples_per_second": 611.776,
-      "eval_steps_per_second": 12.828,
+      "eval_accuracy": 0.8764516129032258,
+      "eval_loss": 0.06577406078577042,
+      "eval_runtime": 5.6348,
+      "eval_samples_per_second": 550.155,
+      "eval_steps_per_second": 11.535,
       "step": 954
     },
     {
       "epoch": 3.1446540880503147,
-      "grad_norm": 0.4341842532157898,
-      "learning_rate": 1.371069182389937e-05,
-      "loss": 0.1085,
+      "grad_norm": 0.44873112440109253,
+      "learning_rate": 1.2138364779874214e-05,
+      "loss": 0.1127,
       "step": 1000
     },
     {
       "epoch": 4.0,
-      "eval_accuracy": 0.9006451612903226,
-      "eval_loss": 0.048103515058755875,
-      "eval_runtime": 5.1362,
-      "eval_samples_per_second": 603.564,
-      "eval_steps_per_second": 12.655,
+      "eval_accuracy": 0.8964516129032258,
+      "eval_loss": 0.05149334296584129,
+      "eval_runtime": 5.7578,
+      "eval_samples_per_second": 538.401,
+      "eval_steps_per_second": 11.289,
       "step": 1272
     },
     {
       "epoch": 4.716981132075472,
-      "grad_norm": 0.29487845301628113,
-      "learning_rate": 1.0566037735849058e-05,
-      "loss": 0.069,
+      "grad_norm": 0.30973368883132935,
+      "learning_rate": 8.207547169811321e-06,
+      "loss": 0.0733,
       "step": 1500
     },
     {
       "epoch": 5.0,
-      "eval_accuracy": 0.9093548387096774,
-      "eval_loss": 0.03922608867287636,
-      "eval_runtime": 5.2603,
-      "eval_samples_per_second": 589.32,
-      "eval_steps_per_second": 12.357,
+      "eval_accuracy": 0.9051612903225806,
+      "eval_loss": 0.04290274158120155,
+      "eval_runtime": 5.9031,
+      "eval_samples_per_second": 525.15,
+      "eval_steps_per_second": 11.011,
       "step": 1590
     },
     {
       "epoch": 6.0,
-      "eval_accuracy": 0.912258064516129,
-      "eval_loss": 0.03417789563536644,
-      "eval_runtime": 5.0804,
-      "eval_samples_per_second": 610.188,
-      "eval_steps_per_second": 12.794,
+      "eval_accuracy": 0.9093548387096774,
+      "eval_loss": 0.038180265575647354,
+      "eval_runtime": 5.6503,
+      "eval_samples_per_second": 548.642,
+      "eval_steps_per_second": 11.504,
       "step": 1908
     },
     {
       "epoch": 6.289308176100629,
-      "grad_norm": 0.2546312212944031,
-      "learning_rate": 7.421383647798742e-06,
-      "loss": 0.0537,
+      "grad_norm": 0.27582159638404846,
+      "learning_rate": 4.276729559748428e-06,
+      "loss": 0.0587,
       "step": 2000
     }
   ],
   "logging_steps": 500,
-  "max_steps": 3180,
+  "max_steps": 2544,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 10,
+  "num_train_epochs": 8,
   "save_steps": 500,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -112,8 +112,8 @@
   "train_batch_size": 48,
   "trial_name": null,
   "trial_params": {
-    "alpha": 0.5365842802514643,
-    "num_train_epochs": 10,
-    "temperature": 14
+    "alpha": 0.819089504077056,
+    "num_train_epochs": 8,
+    "temperature": 12
   }
 }

run-1/checkpoint-2000/training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aff76e4271bd0bf9763ec56f904c0eab2f55fe172f25f099aa8c57c7f7749007
+oid sha256:3ad3761ee6c049b5f54cddb21d4403152bfaafdbfc2a13b71993419e99660838
 size 5240

runs/Oct11_19-31-02_821d3e23518d/events.out.tfevents.1728677628.821d3e23518d.3094.2

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6f994ca38535f739602e335d99d456f787cf08c312db3c4c1a2d64e0685d5087
-size 14689
+oid sha256:db92982d92197dc8e5662c5347cea364558ebe2f532204586fc4df9cf19e8042
+size 15546