Upload folder using huggingface_hub

adapter_config.json

@@ -16,13 +16,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "o_proj",
-    "k_proj",
-    "lm_head",
-    "up_proj",
     "q_proj",
+    "up_proj",
     "down_proj",
     "v_proj",
+    "o_proj",
+    "k_proj",
+    "lm_head",
     "gate_proj"
   ],
   "task_type": "CAUSAL_LM"

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dc52fbefd46700689b7d0b673a8aa5a33d47636420909c812343f097e0a1e253
+oid sha256:8e727969d516177672a61cec24829444b0db62f1a572eb53f76d90d7a2d3db94
 size 85202645

optimizer.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0431726d32c5ac3b1596f40a867ca5ddc342e80e1cde50abc24c743dccbe7ac4
+oid sha256:3aa6a9a2b8d22e58fbb50cc8102365abaee7c846fe3107cd89f2d907cf731229
 size 43126695

rng_state.pth

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:84c0b67b7cc3e4056f21ba7518fc7f338ad65ec2a864cb3572aab3249bf978ea
+oid sha256:67c5005b440f7428a823efa4759cde41dff4f08d3c6e2dea0e2f0eae808d5558
 size 14575

scheduler.pt

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c443d7174c1d7f630d0eb49b5f8e7a4aa613b18d65bcd0dcb596ded57d692b53
+oid sha256:53a98c1866e09b3fe42db209a712ce7090cb5c9a6300901b3c6aa7e319453c22
 size 627

trainer_state.json

@@ -1,9 +1,9 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 1.075794621026895,
+  "epoch": 0.970873786407767,
   "eval_steps": 20,
-  "global_step": 440,
+  "global_step": 400,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
@@ -11,317 +11,289 @@
     {
       "epoch": 0.05,
       "learning_rate": 1.9393939393939395e-05,
-      "loss": 1.8375,
+      "loss": 1.8315,
       "step": 20
     },
     {
       "epoch": 0.05,
-      "eval_loss": 1.7964328527450562,
-      "eval_runtime": 182.6702,
-      "eval_samples_per_second": 1.943,
+      "eval_loss": 1.8022956848144531,
+      "eval_runtime": 192.4256,
+      "eval_samples_per_second": 1.949,
       "eval_steps_per_second": 0.197,
       "step": 20
     },
     {
       "epoch": 0.1,
       "learning_rate": 1.8585858585858588e-05,
-      "loss": 1.7843,
+      "loss": 1.7689,
       "step": 40
     },
     {
       "epoch": 0.1,
-      "eval_loss": 1.7467044591903687,
-      "eval_runtime": 182.464,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "eval_loss": 1.7563962936401367,
+      "eval_runtime": 192.2202,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "step": 40
     },
     {
       "epoch": 0.15,
       "learning_rate": 1.7777777777777777e-05,
-      "loss": 1.7024,
+      "loss": 1.7292,
       "step": 60
     },
     {
       "epoch": 0.15,
-      "eval_loss": 1.715632438659668,
-      "eval_runtime": 182.468,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "eval_loss": 1.7281174659729004,
+      "eval_runtime": 192.2275,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "step": 60
     },
     {
-      "epoch": 0.2,
+      "epoch": 0.19,
       "learning_rate": 1.6969696969696972e-05,
-      "loss": 1.7213,
+      "loss": 1.7028,
       "step": 80
     },
     {
-      "epoch": 0.2,
-      "eval_loss": 1.6986892223358154,
-      "eval_runtime": 182.4667,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "epoch": 0.19,
+      "eval_loss": 1.7117385864257812,
+      "eval_runtime": 192.209,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "step": 80
     },
     {
       "epoch": 0.24,
       "learning_rate": 1.616161616161616e-05,
-      "loss": 1.7345,
+      "loss": 1.6939,
       "step": 100
     },
     {
       "epoch": 0.24,
-      "eval_loss": 1.6884088516235352,
-      "eval_runtime": 182.474,
-      "eval_samples_per_second": 1.945,
-      "eval_steps_per_second": 0.197,
+      "eval_loss": 1.701405644416809,
+      "eval_runtime": 192.2738,
+      "eval_samples_per_second": 1.95,
+      "eval_steps_per_second": 0.198,
       "step": 100
     },
     {
       "epoch": 0.29,
       "learning_rate": 1.5353535353535354e-05,
-      "loss": 1.7001,
+      "loss": 1.6889,
       "step": 120
     },
     {
       "epoch": 0.29,
-      "eval_loss": 1.680200219154358,
-      "eval_runtime": 182.4652,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "eval_loss": 1.6941999197006226,
+      "eval_runtime": 192.278,
+      "eval_samples_per_second": 1.95,
+      "eval_steps_per_second": 0.198,
       "step": 120
     },
     {
       "epoch": 0.34,
       "learning_rate": 1.4545454545454546e-05,
-      "loss": 1.6955,
+      "loss": 1.6827,
       "step": 140
     },
     {
       "epoch": 0.34,
-      "eval_loss": 1.67453134059906,
-      "eval_runtime": 182.4656,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "eval_loss": 1.6888084411621094,
+      "eval_runtime": 192.3059,
+      "eval_samples_per_second": 1.95,
+      "eval_steps_per_second": 0.198,
       "step": 140
     },
     {
       "epoch": 0.39,
       "learning_rate": 1.3737373737373739e-05,
-      "loss": 1.673,
+      "loss": 1.6611,
       "step": 160
     },
     {
       "epoch": 0.39,
-      "eval_loss": 1.669642686843872,
-      "eval_runtime": 182.4712,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "eval_loss": 1.6843013763427734,
+      "eval_runtime": 192.286,
+      "eval_samples_per_second": 1.95,
+      "eval_steps_per_second": 0.198,
       "step": 160
     },
     {
       "epoch": 0.44,
       "learning_rate": 1.2929292929292931e-05,
-      "loss": 1.6703,
+      "loss": 1.6672,
       "step": 180
     },
     {
       "epoch": 0.44,
-      "eval_loss": 1.6656941175460815,
-      "eval_runtime": 182.4669,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "eval_loss": 1.6804944276809692,
+      "eval_runtime": 192.2525,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "step": 180
     },
     {
       "epoch": 0.49,
       "learning_rate": 1.2121212121212122e-05,
-      "loss": 1.6827,
+      "loss": 1.6435,
       "step": 200
     },
     {
       "epoch": 0.49,
-      "eval_loss": 1.662445306777954,
-      "eval_runtime": 182.4652,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "eval_loss": 1.676919937133789,
+      "eval_runtime": 192.2677,
+      "eval_samples_per_second": 1.95,
+      "eval_steps_per_second": 0.198,
       "step": 200
     },
     {
-      "epoch": 0.54,
+      "epoch": 0.53,
       "learning_rate": 1.1313131313131314e-05,
-      "loss": 1.6761,
+      "loss": 1.6804,
       "step": 220
     },
     {
-      "epoch": 0.54,
-      "eval_loss": 1.66029691696167,
-      "eval_runtime": 182.4637,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "epoch": 0.53,
+      "eval_loss": 1.6742541790008545,
+      "eval_runtime": 192.263,
+      "eval_samples_per_second": 1.95,
+      "eval_steps_per_second": 0.198,
       "step": 220
     },
     {
-      "epoch": 0.59,
+      "epoch": 0.58,
       "learning_rate": 1.0505050505050507e-05,
-      "loss": 1.6584,
+      "loss": 1.6542,
       "step": 240
     },
     {
-      "epoch": 0.59,
-      "eval_loss": 1.6578339338302612,
-      "eval_runtime": 182.4676,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "epoch": 0.58,
+      "eval_loss": 1.6716855764389038,
+      "eval_runtime": 192.2148,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "step": 240
     },
     {
-      "epoch": 0.64,
+      "epoch": 0.63,
       "learning_rate": 9.696969696969698e-06,
-      "loss": 1.628,
+      "loss": 1.6417,
       "step": 260
     },
     {
-      "epoch": 0.64,
-      "eval_loss": 1.6552883386611938,
-      "eval_runtime": 182.4666,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "epoch": 0.63,
+      "eval_loss": 1.6693024635314941,
+      "eval_runtime": 192.2366,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "step": 260
     },
     {
       "epoch": 0.68,
       "learning_rate": 8.888888888888888e-06,
-      "loss": 1.6473,
+      "loss": 1.6895,
       "step": 280
     },
     {
       "epoch": 0.68,
-      "eval_loss": 1.652998924255371,
-      "eval_runtime": 182.4572,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "eval_loss": 1.667194128036499,
+      "eval_runtime": 192.2439,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "step": 280
     },
     {
       "epoch": 0.73,
       "learning_rate": 8.08080808080808e-06,
-      "loss": 1.6428,
+      "loss": 1.6637,
       "step": 300
     },
     {
       "epoch": 0.73,
-      "eval_loss": 1.6517064571380615,
-      "eval_runtime": 182.449,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
+      "eval_loss": 1.6655081510543823,
+      "eval_runtime": 192.2484,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "step": 300
     },
     {
       "epoch": 0.78,
       "learning_rate": 7.272727272727273e-06,
-      "loss": 1.6488,
+      "loss": 1.6537,
       "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,
+      "eval_loss": 1.663667917251587,
+      "eval_runtime": 192.2591,
+      "eval_samples_per_second": 1.95,
+      "eval_steps_per_second": 0.198,
       "step": 320
     },
     {
       "epoch": 0.83,
       "learning_rate": 6.464646464646466e-06,
-      "loss": 1.6577,
+      "loss": 1.6489,
       "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,
+      "eval_loss": 1.6618822813034058,
+      "eval_runtime": 192.2528,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "step": 340
     },
     {
-      "epoch": 0.88,
+      "epoch": 0.87,
       "learning_rate": 5.656565656565657e-06,
-      "loss": 1.6281,
+      "loss": 1.6555,
       "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,
+      "epoch": 0.87,
+      "eval_loss": 1.6608922481536865,
+      "eval_runtime": 192.2301,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "step": 360
     },
     {
-      "epoch": 0.93,
+      "epoch": 0.92,
       "learning_rate": 4.848484848484849e-06,
-      "loss": 1.63,
+      "loss": 1.6515,
       "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,
+      "epoch": 0.92,
+      "eval_loss": 1.6594582796096802,
+      "eval_runtime": 192.2624,
+      "eval_samples_per_second": 1.95,
+      "eval_steps_per_second": 0.198,
       "step": 380
     },
     {
-      "epoch": 0.98,
+      "epoch": 0.97,
       "learning_rate": 4.04040404040404e-06,
-      "loss": 1.6506,
+      "loss": 1.6337,
       "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,
+      "epoch": 0.97,
+      "eval_loss": 1.658281922340393,
+      "eval_runtime": 192.2205,
+      "eval_samples_per_second": 1.951,
+      "eval_steps_per_second": 0.198,
       "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
-    },
-    {
-      "epoch": 1.08,
-      "learning_rate": 2.4242424242424244e-06,
-      "loss": 1.6172,
-      "step": 440
-    },
-    {
-      "epoch": 1.08,
-      "eval_loss": 1.64118492603302,
-      "eval_runtime": 182.448,
-      "eval_samples_per_second": 1.946,
-      "eval_steps_per_second": 0.197,
-      "step": 440
     }
   ],
   "logging_steps": 20,
   "max_steps": 500,
   "num_train_epochs": 2,
   "save_steps": 20,
-  "total_flos": 3.0839473559096525e+17,
+  "total_flos": 2.804385205714944e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:40212c95644353b719856f60ff997f45c029dc4cd51992283313f20a545f87ba
+oid sha256:209313c0c53e274af26a94c5922077d921b510aefc7b6b3ffbd52fda65f43f68
 size 4027