Upload folder using huggingface_hub

adapter_config.json

@@ -17,13 +17,13 @@
   "revision": null,
   "target_modules": [
     "o_proj",
-    "down_proj",
-    "lm_head",
     "k_proj",
-    "gate_proj",
-    "q_proj",
+    "lm_head",
     "up_proj",
-    "v_proj"
+    "q_proj",
+    "down_proj",
+    "v_proj",
+    "gate_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b83d77e83ea6b18e0e06dc818c58bbc3ee321e8a82c3c7fccd94a53fdb33d425
+oid sha256:96123af3142e2ede005f39709f5e9c543d74e44ab1f7ebb91fa1400597e1515a
 size 85202645

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:05ffc581f434ec46b5867dd751083c6a73fc0ec7c408e83cb86dc20aecad0189
+oid sha256:27885c12d024b75dad8461c6579d9de0b4f497f1dd9792a76b76483280a87c8e
 size 43126695

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b59c8847e1c811dd88435edba58f2aa65849a7cef6b402bcc17be8208e09b064
+oid sha256:7525d48cc3a7c99ecb0dbaeb631eec5f1b606ee0e55d426ac1d22bf8f04c1c75
 size 14575

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44b355aaccef3913a5e9bba4d219940691ccdfd290aa5455809f4abfdaba7484
+oid sha256:eabe51737cbdb0ae1ea4e7d6d3cf1dc8c428d8543c2b280832dede334ba1fde1
 size 627

trainer_state.json

@@ -1,229 +1,313 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 0.6525285481239804,
+  "epoch": 1.0268948655256724,
   "eval_steps": 20,
-  "global_step": 300,
+  "global_step": 420,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.04,
-      "learning_rate": 1.969849246231156e-05,
-      "loss": 1.7448,
+      "epoch": 0.05,
+      "learning_rate": 1.9393939393939395e-05,
+      "loss": 1.8375,
       "step": 20
     },
     {
-      "epoch": 0.04,
-      "eval_loss": 1.6862075328826904,
-      "eval_runtime": 219.9041,
-      "eval_samples_per_second": 1.964,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.05,
+      "eval_loss": 1.7964328527450562,
+      "eval_runtime": 182.6702,
+      "eval_samples_per_second": 1.943,
+      "eval_steps_per_second": 0.197,
       "step": 20
     },
     {
-      "epoch": 0.09,
-      "learning_rate": 1.9296482412060304e-05,
-      "loss": 1.6491,
+      "epoch": 0.1,
+      "learning_rate": 1.8585858585858588e-05,
+      "loss": 1.7843,
       "step": 40
     },
     {
-      "epoch": 0.09,
-      "eval_loss": 1.6274302005767822,
-      "eval_runtime": 219.8105,
-      "eval_samples_per_second": 1.965,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.1,
+      "eval_loss": 1.7467044591903687,
+      "eval_runtime": 182.464,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 40
     },
     {
-      "epoch": 0.13,
-      "learning_rate": 1.8894472361809046e-05,
-      "loss": 1.6107,
+      "epoch": 0.15,
+      "learning_rate": 1.7777777777777777e-05,
+      "loss": 1.7024,
       "step": 60
     },
     {
-      "epoch": 0.13,
-      "eval_loss": 1.5971616506576538,
-      "eval_runtime": 219.7599,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.15,
+      "eval_loss": 1.715632438659668,
+      "eval_runtime": 182.468,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 60
     },
     {
-      "epoch": 0.17,
-      "learning_rate": 1.8492462311557792e-05,
-      "loss": 1.6148,
+      "epoch": 0.2,
+      "learning_rate": 1.6969696969696972e-05,
+      "loss": 1.7213,
       "step": 80
     },
     {
-      "epoch": 0.17,
-      "eval_loss": 1.5817948579788208,
-      "eval_runtime": 219.7888,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.2,
+      "eval_loss": 1.6986892223358154,
+      "eval_runtime": 182.4667,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 80
     },
     {
-      "epoch": 0.22,
-      "learning_rate": 1.8090452261306535e-05,
-      "loss": 1.5727,
+      "epoch": 0.24,
+      "learning_rate": 1.616161616161616e-05,
+      "loss": 1.7345,
       "step": 100
     },
     {
-      "epoch": 0.22,
-      "eval_loss": 1.5724695920944214,
-      "eval_runtime": 219.7928,
-      "eval_samples_per_second": 1.965,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.24,
+      "eval_loss": 1.6884088516235352,
+      "eval_runtime": 182.474,
+      "eval_samples_per_second": 1.945,
+      "eval_steps_per_second": 0.197,
       "step": 100
     },
     {
-      "epoch": 0.26,
-      "learning_rate": 1.768844221105528e-05,
-      "loss": 1.5578,
+      "epoch": 0.29,
+      "learning_rate": 1.5353535353535354e-05,
+      "loss": 1.7001,
       "step": 120
     },
     {
-      "epoch": 0.26,
-      "eval_loss": 1.5660758018493652,
-      "eval_runtime": 219.7399,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.29,
+      "eval_loss": 1.680200219154358,
+      "eval_runtime": 182.4652,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 120
     },
     {
-      "epoch": 0.3,
-      "learning_rate": 1.728643216080402e-05,
-      "loss": 1.5592,
+      "epoch": 0.34,
+      "learning_rate": 1.4545454545454546e-05,
+      "loss": 1.6955,
       "step": 140
     },
     {
-      "epoch": 0.3,
-      "eval_loss": 1.5614327192306519,
-      "eval_runtime": 219.7821,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.34,
+      "eval_loss": 1.67453134059906,
+      "eval_runtime": 182.4656,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 140
     },
     {
-      "epoch": 0.35,
-      "learning_rate": 1.6884422110552766e-05,
-      "loss": 1.5719,
+      "epoch": 0.39,
+      "learning_rate": 1.3737373737373739e-05,
+      "loss": 1.673,
       "step": 160
     },
     {
-      "epoch": 0.35,
-      "eval_loss": 1.5577027797698975,
-      "eval_runtime": 219.8353,
-      "eval_samples_per_second": 1.965,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.39,
+      "eval_loss": 1.669642686843872,
+      "eval_runtime": 182.4712,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 160
     },
     {
-      "epoch": 0.39,
-      "learning_rate": 1.6482412060301508e-05,
-      "loss": 1.5437,
+      "epoch": 0.44,
+      "learning_rate": 1.2929292929292931e-05,
+      "loss": 1.6703,
       "step": 180
     },
     {
-      "epoch": 0.39,
-      "eval_loss": 1.5545556545257568,
-      "eval_runtime": 219.7845,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.44,
+      "eval_loss": 1.6656941175460815,
+      "eval_runtime": 182.4669,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 180
     },
     {
-      "epoch": 0.44,
-      "learning_rate": 1.6080402010050254e-05,
-      "loss": 1.5645,
+      "epoch": 0.49,
+      "learning_rate": 1.2121212121212122e-05,
+      "loss": 1.6827,
       "step": 200
     },
     {
-      "epoch": 0.44,
-      "eval_loss": 1.5518943071365356,
-      "eval_runtime": 219.742,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.49,
+      "eval_loss": 1.662445306777954,
+      "eval_runtime": 182.4652,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 200
     },
     {
-      "epoch": 0.48,
-      "learning_rate": 1.5678391959798997e-05,
-      "loss": 1.5687,
+      "epoch": 0.54,
+      "learning_rate": 1.1313131313131314e-05,
+      "loss": 1.6761,
       "step": 220
     },
     {
-      "epoch": 0.48,
-      "eval_loss": 1.549355149269104,
-      "eval_runtime": 219.7771,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.54,
+      "eval_loss": 1.66029691696167,
+      "eval_runtime": 182.4637,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 220
     },
     {
-      "epoch": 0.52,
-      "learning_rate": 1.527638190954774e-05,
-      "loss": 1.5692,
+      "epoch": 0.59,
+      "learning_rate": 1.0505050505050507e-05,
+      "loss": 1.6584,
       "step": 240
     },
     {
-      "epoch": 0.52,
-      "eval_loss": 1.5471032857894897,
-      "eval_runtime": 219.7527,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.59,
+      "eval_loss": 1.6578339338302612,
+      "eval_runtime": 182.4676,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 240
     },
     {
-      "epoch": 0.57,
-      "learning_rate": 1.4874371859296483e-05,
-      "loss": 1.5821,
+      "epoch": 0.64,
+      "learning_rate": 9.696969696969698e-06,
+      "loss": 1.628,
       "step": 260
     },
     {
-      "epoch": 0.57,
-      "eval_loss": 1.5451416969299316,
-      "eval_runtime": 219.7519,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.64,
+      "eval_loss": 1.6552883386611938,
+      "eval_runtime": 182.4666,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 260
     },
     {
-      "epoch": 0.61,
-      "learning_rate": 1.4472361809045228e-05,
-      "loss": 1.54,
+      "epoch": 0.68,
+      "learning_rate": 8.888888888888888e-06,
+      "loss": 1.6473,
       "step": 280
     },
     {
-      "epoch": 0.61,
-      "eval_loss": 1.5428489446640015,
-      "eval_runtime": 219.7649,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.68,
+      "eval_loss": 1.652998924255371,
+      "eval_runtime": 182.4572,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 280
     },
     {
-      "epoch": 0.65,
-      "learning_rate": 1.4070351758793972e-05,
-      "loss": 1.5771,
+      "epoch": 0.73,
+      "learning_rate": 8.08080808080808e-06,
+      "loss": 1.6428,
       "step": 300
     },
     {
-      "epoch": 0.65,
-      "eval_loss": 1.5406657457351685,
-      "eval_runtime": 219.7434,
-      "eval_samples_per_second": 1.966,
-      "eval_steps_per_second": 0.2,
+      "epoch": 0.73,
+      "eval_loss": 1.6517064571380615,
+      "eval_runtime": 182.449,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
       "step": 300
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 7.272727272727273e-06,
+      "loss": 1.6488,
+      "step": 320
+    },
+    {
+      "epoch": 0.78,
+      "eval_loss": 1.6494264602661133,
+      "eval_runtime": 182.4576,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
+      "step": 320
+    },
+    {
+      "epoch": 0.83,
+      "learning_rate": 6.464646464646466e-06,
+      "loss": 1.6577,
+      "step": 340
+    },
+    {
+      "epoch": 0.83,
+      "eval_loss": 1.6474355459213257,
+      "eval_runtime": 182.451,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
+      "step": 340
+    },
+    {
+      "epoch": 0.88,
+      "learning_rate": 5.656565656565657e-06,
+      "loss": 1.6281,
+      "step": 360
+    },
+    {
+      "epoch": 0.88,
+      "eval_loss": 1.6457959413528442,
+      "eval_runtime": 182.4538,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
+      "step": 360
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 4.848484848484849e-06,
+      "loss": 1.63,
+      "step": 380
+    },
+    {
+      "epoch": 0.93,
+      "eval_loss": 1.6441447734832764,
+      "eval_runtime": 182.4409,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
+      "step": 380
+    },
+    {
+      "epoch": 0.98,
+      "learning_rate": 4.04040404040404e-06,
+      "loss": 1.6506,
+      "step": 400
+    },
+    {
+      "epoch": 0.98,
+      "eval_loss": 1.642810344696045,
+      "eval_runtime": 182.4488,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
+      "step": 400
+    },
+    {
+      "epoch": 1.03,
+      "learning_rate": 3.232323232323233e-06,
+      "loss": 1.6212,
+      "step": 420
+    },
+    {
+      "epoch": 1.03,
+      "eval_loss": 1.642095685005188,
+      "eval_runtime": 182.4505,
+      "eval_samples_per_second": 1.946,
+      "eval_steps_per_second": 0.197,
+      "step": 420
     }
   ],
   "logging_steps": 20,
-  "max_steps": 1000,
-  "num_train_epochs": 3,
+  "max_steps": 500,
+  "num_train_epochs": 2,
   "save_steps": 20,
-  "total_flos": 2.103288904286208e+17,
+  "total_flos": 2.943728095623905e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:ada01528f9991bb2880bba2862861125a69dda532dd28ec51334f6ef76d52597
+oid sha256:40212c95644353b719856f60ff997f45c029dc4cd51992283313f20a545f87ba
 size 4027