Finetune model with following https://github.com/tatsu-lab/stanford_alpaca#fine-tuning

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e019f0487293f7bdede232d061e4d39315100768ca7a66b14318586e487b623
+oid sha256:a06652ab49c35704c97135a68936d8724b53751af0a922d0df4bf036f272a79f
 size 13835141

pytorch_model-00001-of-00002.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2f559f3e37d5fe208f0841dc756dbf80b149d19a6dc173ab683026def9530d3f
+oid sha256:d31b0599bec571926fcbb4c8d74df6a54b9f856658d578e0bfc91f50f81ef2ef
 size 9979879204

pytorch_model-00002-of-00002.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:81f5ea8e8276cf51325f479e1d9e83c44f1022ad5f3b66fac903b9f336c72509
+oid sha256:f8577b0c87cbe8310a1d2cc2f66be9a972f6c003663538dcb7d57aad5f0353c8
 size 2025273825

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dcd12d19c79af278f28d08dd9e2cd392161951f2f9fa3355e120be3eeb3cee42
+oid sha256:cde37b1d59ec8275aa116b67c6503ca997bd9a90886b106af365cbb8dfef4db9
 size 14575

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:61d142a00f56e370dd08ab7f419d99eb71c744e7cf0ae126b6069a35d7c59ca7
+oid sha256:3e7bdefe584e61a57a129ab2768e96850498503bfc68609b940b74cde526cebd
 size 627

trainer_state.json

@@ -1,36 +1,64 @@
 {
-  "best_metric": 0.252197265625,
-  "best_model_checkpoint": "./models/2023-03-15_1edc3387-3214/checkpoint-5067",
-  "epoch": 1.0,
-  "global_step": 5067,
+  "best_metric": 0.2509765625,
+  "best_model_checkpoint": "./models/2023-03-16_1edc4008-e54e/checkpoint-2534",
+  "epoch": 3.0,
+  "global_step": 7602,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.0,
-      "learning_rate": 1.999868429708572e-05,
-      "loss": 10.1719,
+      "learning_rate": 1.9997369113391214e-05,
+      "loss": 9.8984,
       "step": 1
     },
     {
       "epoch": 1.0,
       "learning_rate": 1.3333333333333333e-05,
-      "loss": 0.2619,
-      "step": 5067
+      "loss": 0.2605,
+      "step": 2534
     },
     {
       "epoch": 1.0,
-      "eval_loss": 0.252197265625,
-      "eval_runtime": 325.5581,
-      "eval_samples_per_second": 13.835,
-      "eval_steps_per_second": 1.729,
-      "step": 5067
+      "eval_loss": 0.2509765625,
+      "eval_runtime": 191.3685,
+      "eval_samples_per_second": 23.536,
+      "eval_steps_per_second": 2.942,
+      "step": 2534
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 6.666666666666667e-06,
+      "loss": 0.1598,
+      "step": 5068
+    },
+    {
+      "epoch": 2.0,
+      "eval_loss": 0.258544921875,
+      "eval_runtime": 159.499,
+      "eval_samples_per_second": 28.238,
+      "eval_steps_per_second": 3.53,
+      "step": 5068
+    },
+    {
+      "epoch": 3.0,
+      "learning_rate": 0.0,
+      "loss": 0.0753,
+      "step": 7602
+    },
+    {
+      "epoch": 3.0,
+      "eval_loss": 0.3056640625,
+      "eval_runtime": 159.3863,
+      "eval_samples_per_second": 28.258,
+      "eval_steps_per_second": 3.532,
+      "step": 7602
     }
   ],
-  "max_steps": 15201,
+  "max_steps": 7602,
   "num_train_epochs": 3,
-  "total_flos": 7.17503717456216e+17,
+  "total_flos": 2.1525111523686482e+18,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:06fdea69e1b755dd45733f4f1ca4d1dcabbbee294889226ecb3c1020cb98963d
+oid sha256:4872b8314368e34757a8c3791ebfcf77a89ff4d439e81f25e5640f42c35bec74
 size 3323