Laurie commited on
Commit
47a9922
1 Parent(s): 982e834

Upload 10 files

Browse files
README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.4.0
adapter_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "baichuan-inc/Baichuan-7B",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 32.0,
11
+ "lora_dropout": 0.1,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 8,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "W_pack"
18
+ ],
19
+ "task_type": "CAUSAL_LM"
20
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:89ecbea566551453abb624441feb0a0e587636b975b586c9673c55cae3e277ca
3
+ size 16799153
all_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99,
3
+ "train_loss": 2.964561536835461,
4
+ "train_runtime": 1710.3688,
5
+ "train_samples_per_second": 5.771,
6
+ "train_steps_per_second": 0.36
7
+ }
finetuning_args.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "finetuning_type": "lora",
3
+ "lora_alpha": 32.0,
4
+ "lora_dropout": 0.1,
5
+ "lora_rank": 8,
6
+ "lora_target": [
7
+ "W_pack"
8
+ ],
9
+ "name_module_trainable": "mlp",
10
+ "num_hidden_layers": 32,
11
+ "num_layer_trainable": 3
12
+ }
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 2.99,
3
+ "train_loss": 2.964561536835461,
4
+ "train_runtime": 1710.3688,
5
+ "train_samples_per_second": 5.771,
6
+ "train_steps_per_second": 0.36
7
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 10, "total_steps": 615, "loss": 3.1898, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.996738892723075e-05, "epoch": 0.05, "percentage": 1.63, "elapsed_time": "0:00:28", "remaining_time": "0:28:38"}
2
+ {"current_steps": 20, "total_steps": 615, "loss": 3.0387, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.986964078748837e-05, "epoch": 0.1, "percentage": 3.25, "elapsed_time": "0:00:56", "remaining_time": "0:27:48"}
3
+ {"current_steps": 30, "total_steps": 615, "loss": 3.1413, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.970701059450872e-05, "epoch": 0.15, "percentage": 4.88, "elapsed_time": "0:01:22", "remaining_time": "0:26:51"}
4
+ {"current_steps": 40, "total_steps": 615, "loss": 3.0543, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9479922631896405e-05, "epoch": 0.19, "percentage": 6.5, "elapsed_time": "0:01:51", "remaining_time": "0:26:40"}
5
+ {"current_steps": 50, "total_steps": 615, "loss": 3.0544, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.918896934621734e-05, "epoch": 0.24, "percentage": 8.13, "elapsed_time": "0:02:19", "remaining_time": "0:26:13"}
6
+ {"current_steps": 60, "total_steps": 615, "loss": 2.9954, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8834909801373264e-05, "epoch": 0.29, "percentage": 9.76, "elapsed_time": "0:02:45", "remaining_time": "0:25:34"}
7
+ {"current_steps": 70, "total_steps": 615, "loss": 3.0829, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.8418667698290696e-05, "epoch": 0.34, "percentage": 11.38, "elapsed_time": "0:03:14", "remaining_time": "0:25:13"}
8
+ {"current_steps": 80, "total_steps": 615, "loss": 3.0368, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.79413289650907e-05, "epoch": 0.39, "percentage": 13.01, "elapsed_time": "0:03:43", "remaining_time": "0:24:55"}
9
+ {"current_steps": 90, "total_steps": 615, "loss": 2.941, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.740413892402639e-05, "epoch": 0.44, "percentage": 14.63, "elapsed_time": "0:04:10", "remaining_time": "0:24:23"}
10
+ {"current_steps": 100, "total_steps": 615, "loss": 2.9253, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.680849904257938e-05, "epoch": 0.49, "percentage": 16.26, "elapsed_time": "0:04:38", "remaining_time": "0:23:56"}
11
+ {"current_steps": 110, "total_steps": 615, "loss": 2.9969, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.615596327719111e-05, "epoch": 0.53, "percentage": 17.89, "elapsed_time": "0:05:06", "remaining_time": "0:23:26"}
12
+ {"current_steps": 120, "total_steps": 615, "loss": 2.9547, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.5448234019167945e-05, "epoch": 0.58, "percentage": 19.51, "elapsed_time": "0:05:33", "remaining_time": "0:22:55"}
13
+ {"current_steps": 130, "total_steps": 615, "loss": 3.021, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.468715765333664e-05, "epoch": 0.63, "percentage": 21.14, "elapsed_time": "0:06:03", "remaining_time": "0:22:35"}
14
+ {"current_steps": 140, "total_steps": 615, "loss": 2.96, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.387471974103713e-05, "epoch": 0.68, "percentage": 22.76, "elapsed_time": "0:06:30", "remaining_time": "0:22:05"}
15
+ {"current_steps": 150, "total_steps": 615, "loss": 3.0919, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.301303984001967e-05, "epoch": 0.73, "percentage": 24.39, "elapsed_time": "0:06:57", "remaining_time": "0:21:34"}
16
+ {"current_steps": 160, "total_steps": 615, "loss": 2.9032, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.210436597476076e-05, "epoch": 0.78, "percentage": 26.02, "elapsed_time": "0:07:23", "remaining_time": "0:21:02"}
17
+ {"current_steps": 170, "total_steps": 615, "loss": 2.9535, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1151068771623866e-05, "epoch": 0.83, "percentage": 27.64, "elapsed_time": "0:07:52", "remaining_time": "0:20:37"}
18
+ {"current_steps": 180, "total_steps": 615, "loss": 2.9935, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.015563527416595e-05, "epoch": 0.87, "percentage": 29.27, "elapsed_time": "0:08:21", "remaining_time": "0:20:12"}
19
+ {"current_steps": 190, "total_steps": 615, "loss": 3.0283, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.9120662454724836e-05, "epoch": 0.92, "percentage": 30.89, "elapsed_time": "0:08:48", "remaining_time": "0:19:42"}
20
+ {"current_steps": 200, "total_steps": 615, "loss": 2.9129, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.8048850439214844e-05, "epoch": 0.97, "percentage": 32.52, "elapsed_time": "0:09:15", "remaining_time": "0:19:13"}
21
+ {"current_steps": 210, "total_steps": 615, "loss": 2.9757, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.694299546280657e-05, "epoch": 1.02, "percentage": 34.15, "elapsed_time": "0:09:45", "remaining_time": "0:18:48"}
22
+ {"current_steps": 220, "total_steps": 615, "loss": 2.9132, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.580598257486867e-05, "epoch": 1.07, "percentage": 35.77, "elapsed_time": "0:10:13", "remaining_time": "0:18:20"}
23
+ {"current_steps": 230, "total_steps": 615, "loss": 2.8876, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.46407781122034e-05, "epoch": 1.12, "percentage": 37.4, "elapsed_time": "0:10:40", "remaining_time": "0:17:52"}
24
+ {"current_steps": 240, "total_steps": 615, "loss": 2.9416, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.3450421960212566e-05, "epoch": 1.17, "percentage": 39.02, "elapsed_time": "0:11:08", "remaining_time": "0:17:24"}
25
+ {"current_steps": 250, "total_steps": 615, "loss": 2.994, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.223801962218372e-05, "epoch": 1.22, "percentage": 40.65, "elapsed_time": "0:11:36", "remaining_time": "0:16:56"}
26
+ {"current_steps": 260, "total_steps": 615, "loss": 2.9024, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.100673411738652e-05, "epoch": 1.26, "percentage": 42.28, "elapsed_time": "0:12:03", "remaining_time": "0:16:28"}
27
+ {"current_steps": 270, "total_steps": 615, "loss": 2.9789, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.975977772911671e-05, "epoch": 1.31, "percentage": 43.9, "elapsed_time": "0:12:31", "remaining_time": "0:16:00"}
28
+ {"current_steps": 280, "total_steps": 615, "loss": 2.9478, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8500403624215734e-05, "epoch": 1.36, "percentage": 45.53, "elapsed_time": "0:13:00", "remaining_time": "0:15:33"}
29
+ {"current_steps": 290, "total_steps": 615, "loss": 2.8663, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.723189736592986e-05, "epoch": 1.41, "percentage": 47.15, "elapsed_time": "0:13:27", "remaining_time": "0:15:04"}
30
+ {"current_steps": 300, "total_steps": 615, "loss": 3.0051, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.595756834225089e-05, "epoch": 1.46, "percentage": 48.78, "elapsed_time": "0:13:55", "remaining_time": "0:14:37"}
31
+ {"current_steps": 310, "total_steps": 615, "loss": 2.8488, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.468074113210066e-05, "epoch": 1.51, "percentage": 50.41, "elapsed_time": "0:14:22", "remaining_time": "0:14:09"}
32
+ {"current_steps": 320, "total_steps": 615, "loss": 2.9825, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.340474683188429e-05, "epoch": 1.56, "percentage": 52.03, "elapsed_time": "0:14:50", "remaining_time": "0:13:40"}
33
+ {"current_steps": 330, "total_steps": 615, "loss": 3.0427, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2132914365039993e-05, "epoch": 1.6, "percentage": 53.66, "elapsed_time": "0:15:18", "remaining_time": "0:13:12"}
34
+ {"current_steps": 340, "total_steps": 615, "loss": 3.0026, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0868561797257878e-05, "epoch": 1.65, "percentage": 55.28, "elapsed_time": "0:15:44", "remaining_time": "0:12:44"}
35
+ {"current_steps": 350, "total_steps": 615, "loss": 2.984, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.961498768002547e-05, "epoch": 1.7, "percentage": 56.91, "elapsed_time": "0:16:11", "remaining_time": "0:12:15"}
36
+ {"current_steps": 360, "total_steps": 615, "loss": 2.9845, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8375462445083464e-05, "epoch": 1.75, "percentage": 58.54, "elapsed_time": "0:16:42", "remaining_time": "0:11:49"}
37
+ {"current_steps": 370, "total_steps": 615, "loss": 2.9139, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.7153219872242727e-05, "epoch": 1.8, "percentage": 60.16, "elapsed_time": "0:17:09", "remaining_time": "0:11:21"}
38
+ {"current_steps": 380, "total_steps": 615, "loss": 3.0234, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5951448652822047e-05, "epoch": 1.85, "percentage": 61.79, "elapsed_time": "0:17:35", "remaining_time": "0:10:52"}
39
+ {"current_steps": 390, "total_steps": 615, "loss": 2.9504, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.4773284070716503e-05, "epoch": 1.9, "percentage": 63.41, "elapsed_time": "0:18:03", "remaining_time": "0:10:25"}
40
+ {"current_steps": 400, "total_steps": 615, "loss": 2.9035, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3621799822799788e-05, "epoch": 1.94, "percentage": 65.04, "elapsed_time": "0:18:32", "remaining_time": "0:09:58"}
41
+ {"current_steps": 410, "total_steps": 615, "loss": 2.9339, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.2500000000000006e-05, "epoch": 1.99, "percentage": 66.67, "elapsed_time": "0:19:00", "remaining_time": "0:09:30"}
42
+ {"current_steps": 420, "total_steps": 615, "loss": 2.9613, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1410811249969475e-05, "epoch": 2.04, "percentage": 68.29, "elapsed_time": "0:19:27", "remaining_time": "0:09:02"}
43
+ {"current_steps": 430, "total_steps": 615, "loss": 2.9762, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.035707514179513e-05, "epoch": 2.09, "percentage": 69.92, "elapsed_time": "0:19:54", "remaining_time": "0:08:34"}
44
+ {"current_steps": 440, "total_steps": 615, "loss": 2.8895, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.341540752669235e-06, "epoch": 2.14, "percentage": 71.54, "elapsed_time": "0:20:22", "remaining_time": "0:08:06"}
45
+ {"current_steps": 450, "total_steps": 615, "loss": 2.9219, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.36685749586087e-06, "epoch": 2.19, "percentage": 73.17, "elapsed_time": "0:20:50", "remaining_time": "0:07:38"}
46
+ {"current_steps": 460, "total_steps": 615, "loss": 2.8863, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.435568208699203e-06, "epoch": 2.24, "percentage": 74.8, "elapsed_time": "0:21:18", "remaining_time": "0:07:10"}
47
+ {"current_steps": 470, "total_steps": 615, "loss": 2.8696, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.55010251860127e-06, "epoch": 2.28, "percentage": 76.42, "elapsed_time": "0:21:48", "remaining_time": "0:06:43"}
48
+ {"current_steps": 480, "total_steps": 615, "loss": 2.8808, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.712770504451426e-06, "epoch": 2.33, "percentage": 78.05, "elapsed_time": "0:22:16", "remaining_time": "0:06:15"}
49
+ {"current_steps": 490, "total_steps": 615, "loss": 2.9685, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.925756669869314e-06, "epoch": 2.38, "percentage": 79.67, "elapsed_time": "0:22:44", "remaining_time": "0:05:48"}
50
+ {"current_steps": 500, "total_steps": 615, "loss": 2.8873, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.19111424408932e-06, "epoch": 2.43, "percentage": 81.3, "elapsed_time": "0:23:10", "remaining_time": "0:05:19"}
51
+ {"current_steps": 510, "total_steps": 615, "loss": 2.9854, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5107598253199758e-06, "epoch": 2.48, "percentage": 82.93, "elapsed_time": "0:23:39", "remaining_time": "0:04:52"}
52
+ {"current_steps": 520, "total_steps": 615, "loss": 2.9411, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8864683805580133e-06, "epoch": 2.53, "percentage": 84.55, "elapsed_time": "0:24:06", "remaining_time": "0:04:24"}
53
+ {"current_steps": 530, "total_steps": 615, "loss": 2.9353, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3198686149022013e-06, "epoch": 2.58, "percentage": 86.18, "elapsed_time": "0:24:32", "remaining_time": "0:03:56"}
54
+ {"current_steps": 540, "total_steps": 615, "loss": 2.9348, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8124387224476347e-06, "epoch": 2.62, "percentage": 87.8, "elapsed_time": "0:25:01", "remaining_time": "0:03:28"}
55
+ {"current_steps": 550, "total_steps": 615, "loss": 2.8754, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.365502529846166e-06, "epoch": 2.67, "percentage": 89.43, "elapsed_time": "0:25:29", "remaining_time": "0:03:00"}
56
+ {"current_steps": 560, "total_steps": 615, "loss": 2.925, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.802260425938099e-07, "epoch": 2.72, "percentage": 91.06, "elapsed_time": "0:25:57", "remaining_time": "0:02:32"}
57
+ {"current_steps": 570, "total_steps": 615, "loss": 2.976, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.576144030555259e-07, "epoch": 2.77, "percentage": 92.68, "elapsed_time": "0:26:26", "remaining_time": "0:02:05"}
58
+ {"current_steps": 580, "total_steps": 615, "loss": 2.913, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.9850926816357157e-07, "epoch": 2.82, "percentage": 94.31, "elapsed_time": "0:26:54", "remaining_time": "0:01:37"}
59
+ {"current_steps": 590, "total_steps": 615, "loss": 2.9485, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0358661363065746e-07, "epoch": 2.87, "percentage": 95.93, "elapsed_time": "0:27:21", "remaining_time": "0:01:09"}
60
+ {"current_steps": 600, "total_steps": 615, "loss": 2.946, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.335497040648898e-08, "epoch": 2.92, "percentage": 97.56, "elapsed_time": "0:27:48", "remaining_time": "0:00:41"}
61
+ {"current_steps": 610, "total_steps": 615, "loss": 2.982, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.154097978591014e-09, "epoch": 2.96, "percentage": 99.19, "elapsed_time": "0:28:16", "remaining_time": "0:00:13"}
62
+ {"current_steps": 615, "total_steps": 615, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 2.99, "percentage": 100.0, "elapsed_time": "0:28:30", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,391 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.98906439854192,
5
+ "global_step": 615,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.05,
12
+ "learning_rate": 4.996738892723075e-05,
13
+ "loss": 3.1898,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 0.1,
18
+ "learning_rate": 4.986964078748837e-05,
19
+ "loss": 3.0387,
20
+ "step": 20
21
+ },
22
+ {
23
+ "epoch": 0.15,
24
+ "learning_rate": 4.970701059450872e-05,
25
+ "loss": 3.1413,
26
+ "step": 30
27
+ },
28
+ {
29
+ "epoch": 0.19,
30
+ "learning_rate": 4.9479922631896405e-05,
31
+ "loss": 3.0543,
32
+ "step": 40
33
+ },
34
+ {
35
+ "epoch": 0.24,
36
+ "learning_rate": 4.918896934621734e-05,
37
+ "loss": 3.0544,
38
+ "step": 50
39
+ },
40
+ {
41
+ "epoch": 0.29,
42
+ "learning_rate": 4.8834909801373264e-05,
43
+ "loss": 2.9954,
44
+ "step": 60
45
+ },
46
+ {
47
+ "epoch": 0.34,
48
+ "learning_rate": 4.8418667698290696e-05,
49
+ "loss": 3.0829,
50
+ "step": 70
51
+ },
52
+ {
53
+ "epoch": 0.39,
54
+ "learning_rate": 4.79413289650907e-05,
55
+ "loss": 3.0368,
56
+ "step": 80
57
+ },
58
+ {
59
+ "epoch": 0.44,
60
+ "learning_rate": 4.740413892402639e-05,
61
+ "loss": 2.941,
62
+ "step": 90
63
+ },
64
+ {
65
+ "epoch": 0.49,
66
+ "learning_rate": 4.680849904257938e-05,
67
+ "loss": 2.9253,
68
+ "step": 100
69
+ },
70
+ {
71
+ "epoch": 0.53,
72
+ "learning_rate": 4.615596327719111e-05,
73
+ "loss": 2.9969,
74
+ "step": 110
75
+ },
76
+ {
77
+ "epoch": 0.58,
78
+ "learning_rate": 4.5448234019167945e-05,
79
+ "loss": 2.9547,
80
+ "step": 120
81
+ },
82
+ {
83
+ "epoch": 0.63,
84
+ "learning_rate": 4.468715765333664e-05,
85
+ "loss": 3.021,
86
+ "step": 130
87
+ },
88
+ {
89
+ "epoch": 0.68,
90
+ "learning_rate": 4.387471974103713e-05,
91
+ "loss": 2.96,
92
+ "step": 140
93
+ },
94
+ {
95
+ "epoch": 0.73,
96
+ "learning_rate": 4.301303984001967e-05,
97
+ "loss": 3.0919,
98
+ "step": 150
99
+ },
100
+ {
101
+ "epoch": 0.78,
102
+ "learning_rate": 4.210436597476076e-05,
103
+ "loss": 2.9032,
104
+ "step": 160
105
+ },
106
+ {
107
+ "epoch": 0.83,
108
+ "learning_rate": 4.1151068771623866e-05,
109
+ "loss": 2.9535,
110
+ "step": 170
111
+ },
112
+ {
113
+ "epoch": 0.87,
114
+ "learning_rate": 4.015563527416595e-05,
115
+ "loss": 2.9935,
116
+ "step": 180
117
+ },
118
+ {
119
+ "epoch": 0.92,
120
+ "learning_rate": 3.9120662454724836e-05,
121
+ "loss": 3.0283,
122
+ "step": 190
123
+ },
124
+ {
125
+ "epoch": 0.97,
126
+ "learning_rate": 3.8048850439214844e-05,
127
+ "loss": 2.9129,
128
+ "step": 200
129
+ },
130
+ {
131
+ "epoch": 1.02,
132
+ "learning_rate": 3.694299546280657e-05,
133
+ "loss": 2.9757,
134
+ "step": 210
135
+ },
136
+ {
137
+ "epoch": 1.07,
138
+ "learning_rate": 3.580598257486867e-05,
139
+ "loss": 2.9132,
140
+ "step": 220
141
+ },
142
+ {
143
+ "epoch": 1.12,
144
+ "learning_rate": 3.46407781122034e-05,
145
+ "loss": 2.8876,
146
+ "step": 230
147
+ },
148
+ {
149
+ "epoch": 1.17,
150
+ "learning_rate": 3.3450421960212566e-05,
151
+ "loss": 2.9416,
152
+ "step": 240
153
+ },
154
+ {
155
+ "epoch": 1.22,
156
+ "learning_rate": 3.223801962218372e-05,
157
+ "loss": 2.994,
158
+ "step": 250
159
+ },
160
+ {
161
+ "epoch": 1.26,
162
+ "learning_rate": 3.100673411738652e-05,
163
+ "loss": 2.9024,
164
+ "step": 260
165
+ },
166
+ {
167
+ "epoch": 1.31,
168
+ "learning_rate": 2.975977772911671e-05,
169
+ "loss": 2.9789,
170
+ "step": 270
171
+ },
172
+ {
173
+ "epoch": 1.36,
174
+ "learning_rate": 2.8500403624215734e-05,
175
+ "loss": 2.9478,
176
+ "step": 280
177
+ },
178
+ {
179
+ "epoch": 1.41,
180
+ "learning_rate": 2.723189736592986e-05,
181
+ "loss": 2.8663,
182
+ "step": 290
183
+ },
184
+ {
185
+ "epoch": 1.46,
186
+ "learning_rate": 2.595756834225089e-05,
187
+ "loss": 3.0051,
188
+ "step": 300
189
+ },
190
+ {
191
+ "epoch": 1.51,
192
+ "learning_rate": 2.468074113210066e-05,
193
+ "loss": 2.8488,
194
+ "step": 310
195
+ },
196
+ {
197
+ "epoch": 1.56,
198
+ "learning_rate": 2.340474683188429e-05,
199
+ "loss": 2.9825,
200
+ "step": 320
201
+ },
202
+ {
203
+ "epoch": 1.6,
204
+ "learning_rate": 2.2132914365039993e-05,
205
+ "loss": 3.0427,
206
+ "step": 330
207
+ },
208
+ {
209
+ "epoch": 1.65,
210
+ "learning_rate": 2.0868561797257878e-05,
211
+ "loss": 3.0026,
212
+ "step": 340
213
+ },
214
+ {
215
+ "epoch": 1.7,
216
+ "learning_rate": 1.961498768002547e-05,
217
+ "loss": 2.984,
218
+ "step": 350
219
+ },
220
+ {
221
+ "epoch": 1.75,
222
+ "learning_rate": 1.8375462445083464e-05,
223
+ "loss": 2.9845,
224
+ "step": 360
225
+ },
226
+ {
227
+ "epoch": 1.8,
228
+ "learning_rate": 1.7153219872242727e-05,
229
+ "loss": 2.9139,
230
+ "step": 370
231
+ },
232
+ {
233
+ "epoch": 1.85,
234
+ "learning_rate": 1.5951448652822047e-05,
235
+ "loss": 3.0234,
236
+ "step": 380
237
+ },
238
+ {
239
+ "epoch": 1.9,
240
+ "learning_rate": 1.4773284070716503e-05,
241
+ "loss": 2.9504,
242
+ "step": 390
243
+ },
244
+ {
245
+ "epoch": 1.94,
246
+ "learning_rate": 1.3621799822799788e-05,
247
+ "loss": 2.9035,
248
+ "step": 400
249
+ },
250
+ {
251
+ "epoch": 1.99,
252
+ "learning_rate": 1.2500000000000006e-05,
253
+ "loss": 2.9339,
254
+ "step": 410
255
+ },
256
+ {
257
+ "epoch": 2.04,
258
+ "learning_rate": 1.1410811249969475e-05,
259
+ "loss": 2.9613,
260
+ "step": 420
261
+ },
262
+ {
263
+ "epoch": 2.09,
264
+ "learning_rate": 1.035707514179513e-05,
265
+ "loss": 2.9762,
266
+ "step": 430
267
+ },
268
+ {
269
+ "epoch": 2.14,
270
+ "learning_rate": 9.341540752669235e-06,
271
+ "loss": 2.8895,
272
+ "step": 440
273
+ },
274
+ {
275
+ "epoch": 2.19,
276
+ "learning_rate": 8.36685749586087e-06,
277
+ "loss": 2.9219,
278
+ "step": 450
279
+ },
280
+ {
281
+ "epoch": 2.24,
282
+ "learning_rate": 7.435568208699203e-06,
283
+ "loss": 2.8863,
284
+ "step": 460
285
+ },
286
+ {
287
+ "epoch": 2.28,
288
+ "learning_rate": 6.55010251860127e-06,
289
+ "loss": 2.8696,
290
+ "step": 470
291
+ },
292
+ {
293
+ "epoch": 2.33,
294
+ "learning_rate": 5.712770504451426e-06,
295
+ "loss": 2.8808,
296
+ "step": 480
297
+ },
298
+ {
299
+ "epoch": 2.38,
300
+ "learning_rate": 4.925756669869314e-06,
301
+ "loss": 2.9685,
302
+ "step": 490
303
+ },
304
+ {
305
+ "epoch": 2.43,
306
+ "learning_rate": 4.19111424408932e-06,
307
+ "loss": 2.8873,
308
+ "step": 500
309
+ },
310
+ {
311
+ "epoch": 2.48,
312
+ "learning_rate": 3.5107598253199758e-06,
313
+ "loss": 2.9854,
314
+ "step": 510
315
+ },
316
+ {
317
+ "epoch": 2.53,
318
+ "learning_rate": 2.8864683805580133e-06,
319
+ "loss": 2.9411,
320
+ "step": 520
321
+ },
322
+ {
323
+ "epoch": 2.58,
324
+ "learning_rate": 2.3198686149022013e-06,
325
+ "loss": 2.9353,
326
+ "step": 530
327
+ },
328
+ {
329
+ "epoch": 2.62,
330
+ "learning_rate": 1.8124387224476347e-06,
331
+ "loss": 2.9348,
332
+ "step": 540
333
+ },
334
+ {
335
+ "epoch": 2.67,
336
+ "learning_rate": 1.365502529846166e-06,
337
+ "loss": 2.8754,
338
+ "step": 550
339
+ },
340
+ {
341
+ "epoch": 2.72,
342
+ "learning_rate": 9.802260425938099e-07,
343
+ "loss": 2.925,
344
+ "step": 560
345
+ },
346
+ {
347
+ "epoch": 2.77,
348
+ "learning_rate": 6.576144030555259e-07,
349
+ "loss": 2.976,
350
+ "step": 570
351
+ },
352
+ {
353
+ "epoch": 2.82,
354
+ "learning_rate": 3.9850926816357157e-07,
355
+ "loss": 2.913,
356
+ "step": 580
357
+ },
358
+ {
359
+ "epoch": 2.87,
360
+ "learning_rate": 2.0358661363065746e-07,
361
+ "loss": 2.9485,
362
+ "step": 590
363
+ },
364
+ {
365
+ "epoch": 2.92,
366
+ "learning_rate": 7.335497040648898e-08,
367
+ "loss": 2.946,
368
+ "step": 600
369
+ },
370
+ {
371
+ "epoch": 2.96,
372
+ "learning_rate": 8.154097978591014e-09,
373
+ "loss": 2.982,
374
+ "step": 610
375
+ },
376
+ {
377
+ "epoch": 2.99,
378
+ "step": 615,
379
+ "total_flos": 7.322328732745728e+16,
380
+ "train_loss": 2.964561536835461,
381
+ "train_runtime": 1710.3688,
382
+ "train_samples_per_second": 5.771,
383
+ "train_steps_per_second": 0.36
384
+ }
385
+ ],
386
+ "max_steps": 615,
387
+ "num_train_epochs": 3,
388
+ "total_flos": 7.322328732745728e+16,
389
+ "trial_name": null,
390
+ "trial_params": null
391
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a6a7accf57470c06ec32db53579d7f0b36ea62a9b1f1250ab49aa0d99087bb3
3
+ size 3325
training_loss.png ADDED