jeromeramos commited on
Commit
4c0d32c
·
verified ·
1 Parent(s): 53c5773

End of training

Browse files
README.md CHANGED
@@ -27,7 +27,7 @@ print(output["generated_text"])
27
 
28
  ## Training procedure
29
 
30
- [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/jerome-ramos-20/huggingface/runs/59giddcb)
31
 
32
 
33
  This model was trained with SFT.
 
27
 
28
  ## Training procedure
29
 
30
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/jerome-ramos-20/huggingface/runs/gciot9b8)
31
 
32
 
33
  This model was trained with SFT.
all_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 0.013845621322256836,
3
- "total_flos": 4.971111595691213e+16,
4
- "train_loss": 16.890113639831544,
5
- "train_runtime": 487.8407,
6
  "train_samples": 46218,
7
- "train_samples_per_second": 1.312,
8
- "train_steps_per_second": 0.02
9
  }
 
1
  {
2
+ "epoch": 1.9982692973347178,
3
+ "total_flos": 7.179552437307441e+18,
4
+ "train_loss": 0.6496460468319974,
5
+ "train_runtime": 14397.0712,
6
  "train_samples": 46218,
7
+ "train_samples_per_second": 6.42,
8
+ "train_steps_per_second": 0.1
9
  }
model-00001-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7653580215afbcfe93ff934d3744298cf021d9eced1ee1bb0c28c4512aa26033
3
  size 4977222960
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e61c525d461465783652367d19b7a6371dd16da255d6c65c5f2f6e605845ce0f
3
  size 4977222960
model-00002-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f822b28a20a52d126f52f097f7ed9bb1bf624b3b0353fa3c4e8c190a99313ed7
3
  size 4999802720
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3c79e8c60bc89be137d79774d8db6184edcb5f10bdc6fbe8bf984b589d23b1bf
3
  size 4999802720
model-00003-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e49f5ac92dcf91c19476489e033570a7d9f241b73f3dbd6d87cb6c27e5a42168
3
  size 4915916176
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d6d6982c6a9581a0a5a58d9b9d9a141c6dc6a1674b25434d085cdbb277af789d
3
  size 4915916176
model-00004-of-00004.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9c3d207013b8bc5b2513767fe60cf1f1a37747843e9b11812c9e2b2688c1e52f
3
  size 1168663096
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63f65d4495451530d3602f76ed69a111b27d1bcfafb15742af2e0a26b1f9a59d
3
  size 1168663096
runs/May03_12-10-17_w-jerom-inter-play-sim-94c6890b9ccf44ea86f033a3db8a5dbd-86twgc7/events.out.tfevents.1746274370.w-jerom-inter-play-sim-94c6890b9ccf44ea86f033a3db8a5dbd-86twgc7.150195.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:355d15cb08412f7c4492595a3360f3599304d67aa63c6753dfe756b2c968e024
3
+ size 67519
tokenizer.json CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b97b77eee18a756e51f2d1b8f463f76706b46ec12fdc443751f99a8ce81bffc0
3
- size 17212268
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:58bfb2b5b5663b192aca10548c3dea97466878edd0da09981ec4724991f204f1
3
+ size 17212169
train_results.json CHANGED
@@ -1,9 +1,9 @@
1
  {
2
- "epoch": 0.013845621322256836,
3
- "total_flos": 4.971111595691213e+16,
4
- "train_loss": 16.890113639831544,
5
- "train_runtime": 487.8407,
6
  "train_samples": 46218,
7
- "train_samples_per_second": 1.312,
8
- "train_steps_per_second": 0.02
9
  }
 
1
  {
2
+ "epoch": 1.9982692973347178,
3
+ "total_flos": 7.179552437307441e+18,
4
+ "train_loss": 0.6496460468319974,
5
+ "train_runtime": 14397.0712,
6
  "train_samples": 46218,
7
+ "train_samples_per_second": 6.42,
8
+ "train_steps_per_second": 0.1
9
  }
trainer_state.json CHANGED
@@ -2,56 +2,2066 @@
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
- "epoch": 0.013845621322256836,
6
  "eval_steps": 500,
7
- "global_step": 10,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
11
  "log_history": [
12
  {
13
  "epoch": 0.0013845621322256837,
14
- "grad_norm": 93.77859497070312,
15
  "learning_rate": 0.0,
16
  "loss": 3.4941,
17
  "step": 1
18
  },
19
  {
20
  "epoch": 0.006922810661128418,
21
- "grad_norm": 341.4436950683594,
22
- "learning_rate": 0.00015000000000000001,
23
- "loss": 19.3195,
24
  "step": 5
25
  },
26
  {
27
  "epoch": 0.013845621322256836,
28
- "grad_norm": 45.13679885864258,
29
- "learning_rate": 6.030737921409169e-06,
30
- "loss": 17.6258,
31
  "step": 10
32
  },
33
  {
34
- "epoch": 0.013845621322256836,
35
- "eval_loss": 10.07286262512207,
36
- "eval_runtime": 77.8792,
37
- "eval_samples_per_second": 26.567,
38
- "eval_steps_per_second": 1.669,
39
- "step": 10
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
40
  },
41
  {
42
- "epoch": 0.013845621322256836,
43
- "step": 10,
44
- "total_flos": 4.971111595691213e+16,
45
- "train_loss": 16.890113639831544,
46
- "train_runtime": 487.8407,
47
- "train_samples_per_second": 1.312,
48
- "train_steps_per_second": 0.02
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
49
  }
50
  ],
51
  "logging_steps": 5,
52
- "max_steps": 10,
53
  "num_input_tokens_seen": 0,
54
- "num_train_epochs": 1,
55
  "save_steps": 500,
56
  "stateful_callbacks": {
57
  "TrainerControl": {
@@ -59,13 +2069,13 @@
59
  "should_epoch_stop": false,
60
  "should_evaluate": false,
61
  "should_log": false,
62
- "should_save": true,
63
- "should_training_stop": true
64
  },
65
  "attributes": {}
66
  }
67
  },
68
- "total_flos": 4.971111595691213e+16,
69
  "train_batch_size": 4,
70
  "trial_name": null,
71
  "trial_params": null
 
2
  "best_global_step": null,
3
  "best_metric": null,
4
  "best_model_checkpoint": null,
5
+ "epoch": 1.9982692973347178,
6
  "eval_steps": 500,
7
+ "global_step": 1444,
8
  "is_hyper_param_search": false,
9
  "is_local_process_zero": true,
10
  "is_world_process_zero": true,
11
  "log_history": [
12
  {
13
  "epoch": 0.0013845621322256837,
14
+ "grad_norm": 93.79446411132812,
15
  "learning_rate": 0.0,
16
  "loss": 3.4941,
17
  "step": 1
18
  },
19
  {
20
  "epoch": 0.006922810661128418,
21
+ "grad_norm": 36.77499008178711,
22
+ "learning_rate": 5.517241379310345e-06,
23
+ "loss": 3.2218,
24
  "step": 5
25
  },
26
  {
27
  "epoch": 0.013845621322256836,
28
+ "grad_norm": 10.594582557678223,
29
+ "learning_rate": 1.2413793103448277e-05,
30
+ "loss": 2.7503,
31
  "step": 10
32
  },
33
  {
34
+ "epoch": 0.020768431983385256,
35
+ "grad_norm": 7.781283855438232,
36
+ "learning_rate": 1.9310344827586207e-05,
37
+ "loss": 2.4468,
38
+ "step": 15
39
+ },
40
+ {
41
+ "epoch": 0.027691242644513673,
42
+ "grad_norm": 7.754919528961182,
43
+ "learning_rate": 2.620689655172414e-05,
44
+ "loss": 2.2012,
45
+ "step": 20
46
+ },
47
+ {
48
+ "epoch": 0.034614053305642094,
49
+ "grad_norm": 7.591794013977051,
50
+ "learning_rate": 3.310344827586207e-05,
51
+ "loss": 1.8914,
52
+ "step": 25
53
+ },
54
+ {
55
+ "epoch": 0.04153686396677051,
56
+ "grad_norm": 7.193952560424805,
57
+ "learning_rate": 4e-05,
58
+ "loss": 1.4744,
59
+ "step": 30
60
+ },
61
+ {
62
+ "epoch": 0.04845967462789893,
63
+ "grad_norm": 3.179903984069824,
64
+ "learning_rate": 4.689655172413793e-05,
65
+ "loss": 1.0679,
66
+ "step": 35
67
+ },
68
+ {
69
+ "epoch": 0.055382485289027346,
70
+ "grad_norm": 2.7340261936187744,
71
+ "learning_rate": 5.379310344827586e-05,
72
+ "loss": 0.9943,
73
+ "step": 40
74
+ },
75
+ {
76
+ "epoch": 0.06230529595015576,
77
+ "grad_norm": 15.587686538696289,
78
+ "learning_rate": 6.068965517241379e-05,
79
+ "loss": 1.3342,
80
+ "step": 45
81
+ },
82
+ {
83
+ "epoch": 0.06922810661128419,
84
+ "grad_norm": 4.922949314117432,
85
+ "learning_rate": 6.758620689655173e-05,
86
+ "loss": 1.1957,
87
+ "step": 50
88
+ },
89
+ {
90
+ "epoch": 0.0761509172724126,
91
+ "grad_norm": 5.149211406707764,
92
+ "learning_rate": 7.448275862068966e-05,
93
+ "loss": 1.2605,
94
+ "step": 55
95
+ },
96
+ {
97
+ "epoch": 0.08307372793354102,
98
+ "grad_norm": 6.597731113433838,
99
+ "learning_rate": 8.137931034482759e-05,
100
+ "loss": 0.9823,
101
+ "step": 60
102
+ },
103
+ {
104
+ "epoch": 0.08999653859466944,
105
+ "grad_norm": 3.9529898166656494,
106
+ "learning_rate": 8.827586206896552e-05,
107
+ "loss": 0.8549,
108
+ "step": 65
109
+ },
110
+ {
111
+ "epoch": 0.09691934925579786,
112
+ "grad_norm": 2.4064829349517822,
113
+ "learning_rate": 9.517241379310345e-05,
114
+ "loss": 0.8041,
115
+ "step": 70
116
+ },
117
+ {
118
+ "epoch": 0.10384215991692627,
119
+ "grad_norm": 1.8010298013687134,
120
+ "learning_rate": 0.0001020689655172414,
121
+ "loss": 0.7536,
122
+ "step": 75
123
+ },
124
+ {
125
+ "epoch": 0.11076497057805469,
126
+ "grad_norm": 3.9770500659942627,
127
+ "learning_rate": 0.00010896551724137931,
128
+ "loss": 0.7651,
129
+ "step": 80
130
+ },
131
+ {
132
+ "epoch": 0.11768778123918311,
133
+ "grad_norm": 2.3516247272491455,
134
+ "learning_rate": 0.00011586206896551725,
135
+ "loss": 0.7936,
136
+ "step": 85
137
+ },
138
+ {
139
+ "epoch": 0.12461059190031153,
140
+ "grad_norm": 1.8583178520202637,
141
+ "learning_rate": 0.00012275862068965518,
142
+ "loss": 0.7787,
143
+ "step": 90
144
+ },
145
+ {
146
+ "epoch": 0.13153340256143994,
147
+ "grad_norm": 6.823867321014404,
148
+ "learning_rate": 0.0001296551724137931,
149
+ "loss": 0.7719,
150
+ "step": 95
151
+ },
152
+ {
153
+ "epoch": 0.13845621322256838,
154
+ "grad_norm": 2.3340225219726562,
155
+ "learning_rate": 0.00013655172413793104,
156
+ "loss": 0.798,
157
+ "step": 100
158
+ },
159
+ {
160
+ "epoch": 0.14537902388369678,
161
+ "grad_norm": 2.410348415374756,
162
+ "learning_rate": 0.00014344827586206896,
163
+ "loss": 0.9031,
164
+ "step": 105
165
+ },
166
+ {
167
+ "epoch": 0.1523018345448252,
168
+ "grad_norm": 2.160402536392212,
169
+ "learning_rate": 0.0001503448275862069,
170
+ "loss": 0.8415,
171
+ "step": 110
172
+ },
173
+ {
174
+ "epoch": 0.1592246452059536,
175
+ "grad_norm": 6.559976100921631,
176
+ "learning_rate": 0.00015724137931034485,
177
+ "loss": 0.8591,
178
+ "step": 115
179
+ },
180
+ {
181
+ "epoch": 0.16614745586708204,
182
+ "grad_norm": 3.4763753414154053,
183
+ "learning_rate": 0.00016413793103448276,
184
+ "loss": 0.8716,
185
+ "step": 120
186
+ },
187
+ {
188
+ "epoch": 0.17307026652821045,
189
+ "grad_norm": 1.728158950805664,
190
+ "learning_rate": 0.0001710344827586207,
191
+ "loss": 0.8675,
192
+ "step": 125
193
+ },
194
+ {
195
+ "epoch": 0.17999307718933888,
196
+ "grad_norm": 3.4520177841186523,
197
+ "learning_rate": 0.00017793103448275862,
198
+ "loss": 0.8199,
199
+ "step": 130
200
+ },
201
+ {
202
+ "epoch": 0.18691588785046728,
203
+ "grad_norm": 38.351226806640625,
204
+ "learning_rate": 0.00018482758620689654,
205
+ "loss": 1.1011,
206
+ "step": 135
207
+ },
208
+ {
209
+ "epoch": 0.19383869851159571,
210
+ "grad_norm": 5.762701988220215,
211
+ "learning_rate": 0.0001917241379310345,
212
+ "loss": 1.3817,
213
+ "step": 140
214
+ },
215
+ {
216
+ "epoch": 0.20076150917272412,
217
+ "grad_norm": 3.3233392238616943,
218
+ "learning_rate": 0.00019862068965517243,
219
+ "loss": 0.9955,
220
+ "step": 145
221
+ },
222
+ {
223
+ "epoch": 0.20768431983385255,
224
+ "grad_norm": 3.346182107925415,
225
+ "learning_rate": 0.0001999953208384105,
226
+ "loss": 0.8073,
227
+ "step": 150
228
+ },
229
+ {
230
+ "epoch": 0.21460713049498095,
231
+ "grad_norm": 2.507021188735962,
232
+ "learning_rate": 0.0001999763124949378,
233
+ "loss": 0.858,
234
+ "step": 155
235
+ },
236
+ {
237
+ "epoch": 0.22152994115610938,
238
+ "grad_norm": 2.7886600494384766,
239
+ "learning_rate": 0.0001999426852993044,
240
+ "loss": 0.864,
241
+ "step": 160
242
+ },
243
+ {
244
+ "epoch": 0.2284527518172378,
245
+ "grad_norm": 2.1663458347320557,
246
+ "learning_rate": 0.00019989444416858502,
247
+ "loss": 0.8071,
248
+ "step": 165
249
+ },
250
+ {
251
+ "epoch": 0.23537556247836622,
252
+ "grad_norm": 1.3868638277053833,
253
+ "learning_rate": 0.0001998315961567502,
254
+ "loss": 0.8137,
255
+ "step": 170
256
+ },
257
+ {
258
+ "epoch": 0.24229837313949462,
259
+ "grad_norm": 1.236806869506836,
260
+ "learning_rate": 0.00019975415045363467,
261
+ "loss": 0.7904,
262
+ "step": 175
263
+ },
264
+ {
265
+ "epoch": 0.24922118380062305,
266
+ "grad_norm": 1.439016342163086,
267
+ "learning_rate": 0.00019966211838359378,
268
+ "loss": 0.8119,
269
+ "step": 180
270
+ },
271
+ {
272
+ "epoch": 0.25614399446175146,
273
+ "grad_norm": 1.086261510848999,
274
+ "learning_rate": 0.00019955551340384743,
275
+ "loss": 0.8023,
276
+ "step": 185
277
+ },
278
+ {
279
+ "epoch": 0.2630668051228799,
280
+ "grad_norm": 1.1318130493164062,
281
+ "learning_rate": 0.00019943435110251248,
282
+ "loss": 0.828,
283
+ "step": 190
284
+ },
285
+ {
286
+ "epoch": 0.2699896157840083,
287
+ "grad_norm": 1.2403993606567383,
288
+ "learning_rate": 0.0001992986491963232,
289
+ "loss": 0.8145,
290
+ "step": 195
291
+ },
292
+ {
293
+ "epoch": 0.27691242644513675,
294
+ "grad_norm": 1.4960097074508667,
295
+ "learning_rate": 0.00019914842752804103,
296
+ "loss": 0.7984,
297
+ "step": 200
298
+ },
299
+ {
300
+ "epoch": 0.2838352371062651,
301
+ "grad_norm": 1.4651333093643188,
302
+ "learning_rate": 0.00019898370806355263,
303
+ "loss": 0.8001,
304
+ "step": 205
305
+ },
306
+ {
307
+ "epoch": 0.29075804776739356,
308
+ "grad_norm": 1.4130685329437256,
309
+ "learning_rate": 0.00019880451488865836,
310
+ "loss": 0.8206,
311
+ "step": 210
312
+ },
313
+ {
314
+ "epoch": 0.297680858428522,
315
+ "grad_norm": 0.9677681922912598,
316
+ "learning_rate": 0.00019861087420555018,
317
+ "loss": 0.7935,
318
+ "step": 215
319
+ },
320
+ {
321
+ "epoch": 0.3046036690896504,
322
+ "grad_norm": 1.11764657497406,
323
+ "learning_rate": 0.0001984028143289803,
324
+ "loss": 0.796,
325
+ "step": 220
326
+ },
327
+ {
328
+ "epoch": 0.3115264797507788,
329
+ "grad_norm": 1.0255619287490845,
330
+ "learning_rate": 0.00019818036568212106,
331
+ "loss": 0.7745,
332
+ "step": 225
333
+ },
334
+ {
335
+ "epoch": 0.3184492904119072,
336
+ "grad_norm": 1.0820256471633911,
337
+ "learning_rate": 0.00019794356079211604,
338
+ "loss": 0.8127,
339
+ "step": 230
340
+ },
341
+ {
342
+ "epoch": 0.32537210107303566,
343
+ "grad_norm": 0.8666492700576782,
344
+ "learning_rate": 0.00019769243428532422,
345
+ "loss": 0.7707,
346
+ "step": 235
347
+ },
348
+ {
349
+ "epoch": 0.3322949117341641,
350
+ "grad_norm": 0.8272375464439392,
351
+ "learning_rate": 0.00019742702288225652,
352
+ "loss": 0.7894,
353
+ "step": 240
354
+ },
355
+ {
356
+ "epoch": 0.33921772239529246,
357
+ "grad_norm": 0.8006950616836548,
358
+ "learning_rate": 0.00019714736539220648,
359
+ "loss": 0.7879,
360
+ "step": 245
361
+ },
362
+ {
363
+ "epoch": 0.3461405330564209,
364
+ "grad_norm": 0.8647785782814026,
365
+ "learning_rate": 0.00019685350270757555,
366
+ "loss": 0.7865,
367
+ "step": 250
368
+ },
369
+ {
370
+ "epoch": 0.3530633437175493,
371
+ "grad_norm": 0.9071713089942932,
372
+ "learning_rate": 0.0001965454777978936,
373
+ "loss": 0.8074,
374
+ "step": 255
375
+ },
376
+ {
377
+ "epoch": 0.35998615437867776,
378
+ "grad_norm": 0.7992457747459412,
379
+ "learning_rate": 0.00019622333570353567,
380
+ "loss": 0.7847,
381
+ "step": 260
382
+ },
383
+ {
384
+ "epoch": 0.36690896503980613,
385
+ "grad_norm": 0.8099973201751709,
386
+ "learning_rate": 0.00019588712352913625,
387
+ "loss": 0.7822,
388
+ "step": 265
389
+ },
390
+ {
391
+ "epoch": 0.37383177570093457,
392
+ "grad_norm": 0.8181321024894714,
393
+ "learning_rate": 0.00019553689043670127,
394
+ "loss": 0.7677,
395
+ "step": 270
396
+ },
397
+ {
398
+ "epoch": 0.380754586362063,
399
+ "grad_norm": 0.7866135835647583,
400
+ "learning_rate": 0.00019517268763841962,
401
+ "loss": 0.7556,
402
+ "step": 275
403
+ },
404
+ {
405
+ "epoch": 0.38767739702319143,
406
+ "grad_norm": 0.7752543687820435,
407
+ "learning_rate": 0.00019479456838917473,
408
+ "loss": 0.7606,
409
+ "step": 280
410
+ },
411
+ {
412
+ "epoch": 0.39460020768431986,
413
+ "grad_norm": 0.6425966024398804,
414
+ "learning_rate": 0.0001944025879787574,
415
+ "loss": 0.7801,
416
+ "step": 285
417
+ },
418
+ {
419
+ "epoch": 0.40152301834544823,
420
+ "grad_norm": 0.8983914256095886,
421
+ "learning_rate": 0.0001939968037237812,
422
+ "loss": 0.7572,
423
+ "step": 290
424
+ },
425
+ {
426
+ "epoch": 0.40844582900657667,
427
+ "grad_norm": 0.8024505376815796,
428
+ "learning_rate": 0.00019357727495930146,
429
+ "loss": 0.7576,
430
+ "step": 295
431
+ },
432
+ {
433
+ "epoch": 0.4153686396677051,
434
+ "grad_norm": 0.7114760279655457,
435
+ "learning_rate": 0.00019314406303013904,
436
+ "loss": 0.7508,
437
+ "step": 300
438
+ },
439
+ {
440
+ "epoch": 0.42229145032883353,
441
+ "grad_norm": 0.741939902305603,
442
+ "learning_rate": 0.00019269723128191048,
443
+ "loss": 0.7478,
444
+ "step": 305
445
+ },
446
+ {
447
+ "epoch": 0.4292142609899619,
448
+ "grad_norm": 0.7591356635093689,
449
+ "learning_rate": 0.00019223684505176517,
450
+ "loss": 0.7443,
451
+ "step": 310
452
+ },
453
+ {
454
+ "epoch": 0.43613707165109034,
455
+ "grad_norm": 0.7672891616821289,
456
+ "learning_rate": 0.00019176297165883165,
457
+ "loss": 0.7468,
458
+ "step": 315
459
+ },
460
+ {
461
+ "epoch": 0.44305988231221877,
462
+ "grad_norm": 0.7053809762001038,
463
+ "learning_rate": 0.00019127568039437406,
464
+ "loss": 0.7497,
465
+ "step": 320
466
+ },
467
+ {
468
+ "epoch": 0.4499826929733472,
469
+ "grad_norm": 0.652579665184021,
470
+ "learning_rate": 0.00019077504251166006,
471
+ "loss": 0.7438,
472
+ "step": 325
473
+ },
474
+ {
475
+ "epoch": 0.4569055036344756,
476
+ "grad_norm": 0.7232092618942261,
477
+ "learning_rate": 0.00019026113121554204,
478
+ "loss": 0.7275,
479
+ "step": 330
480
+ },
481
+ {
482
+ "epoch": 0.463828314295604,
483
+ "grad_norm": 0.6129412055015564,
484
+ "learning_rate": 0.00018973402165175268,
485
+ "loss": 0.7474,
486
+ "step": 335
487
+ },
488
+ {
489
+ "epoch": 0.47075112495673244,
490
+ "grad_norm": 0.6411823034286499,
491
+ "learning_rate": 0.0001891937908959172,
492
+ "loss": 0.746,
493
+ "step": 340
494
+ },
495
+ {
496
+ "epoch": 0.47767393561786087,
497
+ "grad_norm": 0.6721363663673401,
498
+ "learning_rate": 0.00018864051794228282,
499
+ "loss": 0.7314,
500
+ "step": 345
501
+ },
502
+ {
503
+ "epoch": 0.48459674627898924,
504
+ "grad_norm": 0.7300981879234314,
505
+ "learning_rate": 0.00018807428369216822,
506
+ "loss": 0.7244,
507
+ "step": 350
508
+ },
509
+ {
510
+ "epoch": 0.4915195569401177,
511
+ "grad_norm": 0.6554331183433533,
512
+ "learning_rate": 0.00018749517094213375,
513
+ "loss": 0.7297,
514
+ "step": 355
515
+ },
516
+ {
517
+ "epoch": 0.4984423676012461,
518
+ "grad_norm": 0.841881275177002,
519
+ "learning_rate": 0.00018690326437187472,
520
+ "loss": 0.7317,
521
+ "step": 360
522
+ },
523
+ {
524
+ "epoch": 0.5053651782623745,
525
+ "grad_norm": 0.7245016694068909,
526
+ "learning_rate": 0.00018629865053183911,
527
+ "loss": 0.7202,
528
+ "step": 365
529
+ },
530
+ {
531
+ "epoch": 0.5122879889235029,
532
+ "grad_norm": 0.6532626748085022,
533
+ "learning_rate": 0.0001856814178305722,
534
+ "loss": 0.7346,
535
+ "step": 370
536
+ },
537
+ {
538
+ "epoch": 0.5192107995846313,
539
+ "grad_norm": 0.6652960181236267,
540
+ "learning_rate": 0.00018505165652178893,
541
+ "loss": 0.7336,
542
+ "step": 375
543
+ },
544
+ {
545
+ "epoch": 0.5261336102457598,
546
+ "grad_norm": 0.6723728775978088,
547
+ "learning_rate": 0.00018440945869117675,
548
+ "loss": 0.7144,
549
+ "step": 380
550
+ },
551
+ {
552
+ "epoch": 0.5330564209068882,
553
+ "grad_norm": 0.5743719339370728,
554
+ "learning_rate": 0.00018375491824293067,
555
+ "loss": 0.7091,
556
+ "step": 385
557
+ },
558
+ {
559
+ "epoch": 0.5399792315680166,
560
+ "grad_norm": 0.6457469463348389,
561
+ "learning_rate": 0.00018308813088602227,
562
+ "loss": 0.7328,
563
+ "step": 390
564
+ },
565
+ {
566
+ "epoch": 0.5469020422291451,
567
+ "grad_norm": 0.5656632781028748,
568
+ "learning_rate": 0.00018240919412020466,
569
+ "loss": 0.7094,
570
+ "step": 395
571
+ },
572
+ {
573
+ "epoch": 0.5538248528902735,
574
+ "grad_norm": 0.5852614045143127,
575
+ "learning_rate": 0.00018171820722175604,
576
+ "loss": 0.7098,
577
+ "step": 400
578
+ },
579
+ {
580
+ "epoch": 0.5607476635514018,
581
+ "grad_norm": 0.7568480372428894,
582
+ "learning_rate": 0.00018101527122896293,
583
+ "loss": 0.7102,
584
+ "step": 405
585
+ },
586
+ {
587
+ "epoch": 0.5676704742125303,
588
+ "grad_norm": 0.5651789307594299,
589
+ "learning_rate": 0.0001803004889273463,
590
+ "loss": 0.7014,
591
+ "step": 410
592
+ },
593
+ {
594
+ "epoch": 0.5745932848736587,
595
+ "grad_norm": 0.6919078826904297,
596
+ "learning_rate": 0.0001795739648346317,
597
+ "loss": 0.7062,
598
+ "step": 415
599
+ },
600
+ {
601
+ "epoch": 0.5815160955347871,
602
+ "grad_norm": 0.5818287134170532,
603
+ "learning_rate": 0.00017883580518546647,
604
+ "loss": 0.6783,
605
+ "step": 420
606
+ },
607
+ {
608
+ "epoch": 0.5884389061959155,
609
+ "grad_norm": 0.5284450054168701,
610
+ "learning_rate": 0.00017808611791588584,
611
+ "loss": 0.72,
612
+ "step": 425
613
+ },
614
+ {
615
+ "epoch": 0.595361716857044,
616
+ "grad_norm": 0.5624392032623291,
617
+ "learning_rate": 0.00017732501264752996,
618
+ "loss": 0.7046,
619
+ "step": 430
620
+ },
621
+ {
622
+ "epoch": 0.6022845275181724,
623
+ "grad_norm": 0.5067269802093506,
624
+ "learning_rate": 0.00017655260067161503,
625
+ "loss": 0.7072,
626
+ "step": 435
627
+ },
628
+ {
629
+ "epoch": 0.6092073381793008,
630
+ "grad_norm": 0.6225182414054871,
631
+ "learning_rate": 0.00017576899493265954,
632
+ "loss": 0.71,
633
+ "step": 440
634
+ },
635
+ {
636
+ "epoch": 0.6161301488404292,
637
+ "grad_norm": 0.6028428077697754,
638
+ "learning_rate": 0.00017497431001196943,
639
+ "loss": 0.6896,
640
+ "step": 445
641
+ },
642
+ {
643
+ "epoch": 0.6230529595015576,
644
+ "grad_norm": 0.6322434544563293,
645
+ "learning_rate": 0.00017416866211088366,
646
+ "loss": 0.705,
647
+ "step": 450
648
+ },
649
+ {
650
+ "epoch": 0.629975770162686,
651
+ "grad_norm": 0.5573384165763855,
652
+ "learning_rate": 0.00017335216903378267,
653
+ "loss": 0.7015,
654
+ "step": 455
655
+ },
656
+ {
657
+ "epoch": 0.6368985808238145,
658
+ "grad_norm": 0.563373327255249,
659
+ "learning_rate": 0.00017252495017086296,
660
+ "loss": 0.6952,
661
+ "step": 460
662
+ },
663
+ {
664
+ "epoch": 0.6438213914849429,
665
+ "grad_norm": 0.5643554925918579,
666
+ "learning_rate": 0.0001716871264806791,
667
+ "loss": 0.6768,
668
+ "step": 465
669
+ },
670
+ {
671
+ "epoch": 0.6507442021460713,
672
+ "grad_norm": 0.577429473400116,
673
+ "learning_rate": 0.0001708388204724572,
674
+ "loss": 0.6957,
675
+ "step": 470
676
+ },
677
+ {
678
+ "epoch": 0.6576670128071997,
679
+ "grad_norm": 0.5610324740409851,
680
+ "learning_rate": 0.00016998015618818087,
681
+ "loss": 0.6906,
682
+ "step": 475
683
+ },
684
+ {
685
+ "epoch": 0.6645898234683282,
686
+ "grad_norm": 0.5422099828720093,
687
+ "learning_rate": 0.00016911125918445363,
688
+ "loss": 0.6992,
689
+ "step": 480
690
+ },
691
+ {
692
+ "epoch": 0.6715126341294566,
693
+ "grad_norm": 0.5785917639732361,
694
+ "learning_rate": 0.00016823225651413953,
695
+ "loss": 0.7108,
696
+ "step": 485
697
+ },
698
+ {
699
+ "epoch": 0.6784354447905849,
700
+ "grad_norm": 0.5469449758529663,
701
+ "learning_rate": 0.00016734327670778508,
702
+ "loss": 0.6808,
703
+ "step": 490
704
+ },
705
+ {
706
+ "epoch": 0.6853582554517134,
707
+ "grad_norm": 0.5459027886390686,
708
+ "learning_rate": 0.00016644444975482523,
709
+ "loss": 0.6978,
710
+ "step": 495
711
+ },
712
+ {
713
+ "epoch": 0.6922810661128418,
714
+ "grad_norm": 0.6022639274597168,
715
+ "learning_rate": 0.0001655359070845757,
716
+ "loss": 0.694,
717
+ "step": 500
718
+ },
719
+ {
720
+ "epoch": 0.6992038767739702,
721
+ "grad_norm": 0.5332102179527283,
722
+ "learning_rate": 0.0001646177815470151,
723
+ "loss": 0.6771,
724
+ "step": 505
725
+ },
726
+ {
727
+ "epoch": 0.7061266874350987,
728
+ "grad_norm": 0.5422456860542297,
729
+ "learning_rate": 0.00016369020739335924,
730
+ "loss": 0.6731,
731
+ "step": 510
732
+ },
733
+ {
734
+ "epoch": 0.7130494980962271,
735
+ "grad_norm": 0.5522486567497253,
736
+ "learning_rate": 0.00016275332025643028,
737
+ "loss": 0.6817,
738
+ "step": 515
739
+ },
740
+ {
741
+ "epoch": 0.7199723087573555,
742
+ "grad_norm": 0.5140525698661804,
743
+ "learning_rate": 0.00016180725713082434,
744
+ "loss": 0.6799,
745
+ "step": 520
746
+ },
747
+ {
748
+ "epoch": 0.726895119418484,
749
+ "grad_norm": 0.4954339563846588,
750
+ "learning_rate": 0.0001608521563528795,
751
+ "loss": 0.6878,
752
+ "step": 525
753
+ },
754
+ {
755
+ "epoch": 0.7338179300796123,
756
+ "grad_norm": 0.5099174380302429,
757
+ "learning_rate": 0.00015988815758044792,
758
+ "loss": 0.6634,
759
+ "step": 530
760
+ },
761
+ {
762
+ "epoch": 0.7407407407407407,
763
+ "grad_norm": 0.5720632076263428,
764
+ "learning_rate": 0.00015891540177247467,
765
+ "loss": 0.6927,
766
+ "step": 535
767
+ },
768
+ {
769
+ "epoch": 0.7476635514018691,
770
+ "grad_norm": 0.526336133480072,
771
+ "learning_rate": 0.00015793403116838613,
772
+ "loss": 0.6676,
773
+ "step": 540
774
+ },
775
+ {
776
+ "epoch": 0.7545863620629976,
777
+ "grad_norm": 0.660834789276123,
778
+ "learning_rate": 0.00015694418926729146,
779
+ "loss": 0.6853,
780
+ "step": 545
781
+ },
782
+ {
783
+ "epoch": 0.761509172724126,
784
+ "grad_norm": 0.4995917081832886,
785
+ "learning_rate": 0.00015594602080699956,
786
+ "loss": 0.6701,
787
+ "step": 550
788
+ },
789
+ {
790
+ "epoch": 0.7684319833852544,
791
+ "grad_norm": 0.5519081950187683,
792
+ "learning_rate": 0.00015493967174285512,
793
+ "loss": 0.683,
794
+ "step": 555
795
+ },
796
+ {
797
+ "epoch": 0.7753547940463829,
798
+ "grad_norm": 0.5006475448608398,
799
+ "learning_rate": 0.00015392528922639662,
800
+ "loss": 0.6731,
801
+ "step": 560
802
+ },
803
+ {
804
+ "epoch": 0.7822776047075113,
805
+ "grad_norm": 0.5271349549293518,
806
+ "learning_rate": 0.00015290302158383923,
807
+ "loss": 0.6521,
808
+ "step": 565
809
+ },
810
+ {
811
+ "epoch": 0.7892004153686397,
812
+ "grad_norm": 0.5666903853416443,
813
+ "learning_rate": 0.0001518730182943862,
814
+ "loss": 0.6609,
815
+ "step": 570
816
+ },
817
+ {
818
+ "epoch": 0.796123226029768,
819
+ "grad_norm": 0.6733763217926025,
820
+ "learning_rate": 0.0001508354299683715,
821
+ "loss": 0.66,
822
+ "step": 575
823
+ },
824
+ {
825
+ "epoch": 0.8030460366908965,
826
+ "grad_norm": 0.5356613993644714,
827
+ "learning_rate": 0.0001497904083252371,
828
+ "loss": 0.661,
829
+ "step": 580
830
+ },
831
+ {
832
+ "epoch": 0.8099688473520249,
833
+ "grad_norm": 0.49363628029823303,
834
+ "learning_rate": 0.00014873810617134806,
835
+ "loss": 0.678,
836
+ "step": 585
837
+ },
838
+ {
839
+ "epoch": 0.8168916580131533,
840
+ "grad_norm": 0.5436602830886841,
841
+ "learning_rate": 0.00014767867737764876,
842
+ "loss": 0.6767,
843
+ "step": 590
844
+ },
845
+ {
846
+ "epoch": 0.8238144686742818,
847
+ "grad_norm": 0.46172234416007996,
848
+ "learning_rate": 0.0001466122768571633,
849
+ "loss": 0.6446,
850
+ "step": 595
851
+ },
852
+ {
853
+ "epoch": 0.8307372793354102,
854
+ "grad_norm": 0.500989556312561,
855
+ "learning_rate": 0.0001455390605423437,
856
+ "loss": 0.6697,
857
+ "step": 600
858
+ },
859
+ {
860
+ "epoch": 0.8376600899965386,
861
+ "grad_norm": 0.6502693295478821,
862
+ "learning_rate": 0.00014445918536226904,
863
+ "loss": 0.6383,
864
+ "step": 605
865
+ },
866
+ {
867
+ "epoch": 0.8445829006576671,
868
+ "grad_norm": 0.5176218748092651,
869
+ "learning_rate": 0.00014337280921969858,
870
+ "loss": 0.6805,
871
+ "step": 610
872
+ },
873
+ {
874
+ "epoch": 0.8515057113187954,
875
+ "grad_norm": 0.48482829332351685,
876
+ "learning_rate": 0.0001422800909679829,
877
+ "loss": 0.6593,
878
+ "step": 615
879
+ },
880
+ {
881
+ "epoch": 0.8584285219799238,
882
+ "grad_norm": 0.49078136682510376,
883
+ "learning_rate": 0.00014118119038783588,
884
+ "loss": 0.6509,
885
+ "step": 620
886
+ },
887
+ {
888
+ "epoch": 0.8653513326410522,
889
+ "grad_norm": 0.502877414226532,
890
+ "learning_rate": 0.00014007626816397087,
891
+ "loss": 0.6612,
892
+ "step": 625
893
+ },
894
+ {
895
+ "epoch": 0.8722741433021807,
896
+ "grad_norm": 0.5412324666976929,
897
+ "learning_rate": 0.00013896548586160503,
898
+ "loss": 0.6841,
899
+ "step": 630
900
+ },
901
+ {
902
+ "epoch": 0.8791969539633091,
903
+ "grad_norm": 0.5135715007781982,
904
+ "learning_rate": 0.00013784900590283473,
905
+ "loss": 0.659,
906
+ "step": 635
907
+ },
908
+ {
909
+ "epoch": 0.8861197646244375,
910
+ "grad_norm": 0.4901527762413025,
911
+ "learning_rate": 0.00013672699154288569,
912
+ "loss": 0.6462,
913
+ "step": 640
914
+ },
915
+ {
916
+ "epoch": 0.893042575285566,
917
+ "grad_norm": 0.5581095218658447,
918
+ "learning_rate": 0.00013559960684624112,
919
+ "loss": 0.6599,
920
+ "step": 645
921
+ },
922
+ {
923
+ "epoch": 0.8999653859466944,
924
+ "grad_norm": 0.5191010236740112,
925
+ "learning_rate": 0.0001344670166626519,
926
+ "loss": 0.6635,
927
+ "step": 650
928
+ },
929
+ {
930
+ "epoch": 0.9068881966078228,
931
+ "grad_norm": 0.5085604190826416,
932
+ "learning_rate": 0.0001333293866030317,
933
+ "loss": 0.6682,
934
+ "step": 655
935
+ },
936
+ {
937
+ "epoch": 0.9138110072689511,
938
+ "grad_norm": 0.5115968585014343,
939
+ "learning_rate": 0.0001321868830152407,
940
+ "loss": 0.6503,
941
+ "step": 660
942
+ },
943
+ {
944
+ "epoch": 0.9207338179300796,
945
+ "grad_norm": 0.5414242744445801,
946
+ "learning_rate": 0.00013103967295976179,
947
+ "loss": 0.6456,
948
+ "step": 665
949
+ },
950
+ {
951
+ "epoch": 0.927656628591208,
952
+ "grad_norm": 0.44759950041770935,
953
+ "learning_rate": 0.0001298879241852724,
954
+ "loss": 0.6508,
955
+ "step": 670
956
+ },
957
+ {
958
+ "epoch": 0.9345794392523364,
959
+ "grad_norm": 0.5465823411941528,
960
+ "learning_rate": 0.0001287318051041159,
961
+ "loss": 0.64,
962
+ "step": 675
963
+ },
964
+ {
965
+ "epoch": 0.9415022499134649,
966
+ "grad_norm": 0.47527021169662476,
967
+ "learning_rate": 0.00012757148476767553,
968
+ "loss": 0.6464,
969
+ "step": 680
970
+ },
971
+ {
972
+ "epoch": 0.9484250605745933,
973
+ "grad_norm": 0.48905277252197266,
974
+ "learning_rate": 0.0001264071328416555,
975
+ "loss": 0.6365,
976
+ "step": 685
977
+ },
978
+ {
979
+ "epoch": 0.9553478712357217,
980
+ "grad_norm": 0.49631747603416443,
981
+ "learning_rate": 0.00012523891958127168,
982
+ "loss": 0.6665,
983
+ "step": 690
984
+ },
985
+ {
986
+ "epoch": 0.9622706818968502,
987
+ "grad_norm": 0.495635449886322,
988
+ "learning_rate": 0.0001240670158063565,
989
+ "loss": 0.652,
990
+ "step": 695
991
+ },
992
+ {
993
+ "epoch": 0.9691934925579785,
994
+ "grad_norm": 0.48297345638275146,
995
+ "learning_rate": 0.00012289159287638105,
996
+ "loss": 0.6384,
997
+ "step": 700
998
+ },
999
+ {
1000
+ "epoch": 0.9761163032191069,
1001
+ "grad_norm": 0.480582594871521,
1002
+ "learning_rate": 0.00012171282266539845,
1003
+ "loss": 0.6581,
1004
+ "step": 705
1005
+ },
1006
+ {
1007
+ "epoch": 0.9830391138802353,
1008
+ "grad_norm": 0.4751047194004059,
1009
+ "learning_rate": 0.00012053087753691172,
1010
+ "loss": 0.6491,
1011
+ "step": 710
1012
+ },
1013
+ {
1014
+ "epoch": 0.9899619245413638,
1015
+ "grad_norm": 0.4794911742210388,
1016
+ "learning_rate": 0.00011934593031867038,
1017
+ "loss": 0.6523,
1018
+ "step": 715
1019
+ },
1020
+ {
1021
+ "epoch": 0.9968847352024922,
1022
+ "grad_norm": 0.4807172119617462,
1023
+ "learning_rate": 0.00011815815427739911,
1024
+ "loss": 0.6375,
1025
+ "step": 720
1026
+ },
1027
+ {
1028
+ "epoch": 1.0,
1029
+ "eval_loss": 0.635538637638092,
1030
+ "eval_runtime": 79.5539,
1031
+ "eval_samples_per_second": 26.008,
1032
+ "eval_steps_per_second": 1.634,
1033
+ "step": 723
1034
+ },
1035
+ {
1036
+ "epoch": 1.0027691242644514,
1037
+ "grad_norm": 0.48584648966789246,
1038
+ "learning_rate": 0.00011696772309346182,
1039
+ "loss": 0.5824,
1040
+ "step": 725
1041
+ },
1042
+ {
1043
+ "epoch": 1.0096919349255797,
1044
+ "grad_norm": 0.5064235329627991,
1045
+ "learning_rate": 0.0001157748108354659,
1046
+ "loss": 0.5255,
1047
+ "step": 730
1048
+ },
1049
+ {
1050
+ "epoch": 1.0166147455867083,
1051
+ "grad_norm": 0.4342171251773834,
1052
+ "learning_rate": 0.00011457959193480925,
1053
+ "loss": 0.5217,
1054
+ "step": 735
1055
+ },
1056
+ {
1057
+ "epoch": 1.0235375562478366,
1058
+ "grad_norm": 0.4713837206363678,
1059
+ "learning_rate": 0.00011338224116017423,
1060
+ "loss": 0.5235,
1061
+ "step": 740
1062
+ },
1063
+ {
1064
+ "epoch": 1.0304603669089651,
1065
+ "grad_norm": 0.45868873596191406,
1066
+ "learning_rate": 0.00011218293359197267,
1067
+ "loss": 0.5336,
1068
+ "step": 745
1069
+ },
1070
+ {
1071
+ "epoch": 1.0373831775700935,
1072
+ "grad_norm": 0.5150412917137146,
1073
+ "learning_rate": 0.00011098184459674484,
1074
+ "loss": 0.5154,
1075
+ "step": 750
1076
+ },
1077
+ {
1078
+ "epoch": 1.0443059882312218,
1079
+ "grad_norm": 0.44210168719291687,
1080
+ "learning_rate": 0.00010977914980151691,
1081
+ "loss": 0.5144,
1082
+ "step": 755
1083
+ },
1084
+ {
1085
+ "epoch": 1.0512287988923503,
1086
+ "grad_norm": 0.45892542600631714,
1087
+ "learning_rate": 0.00010857502506812029,
1088
+ "loss": 0.49,
1089
+ "step": 760
1090
+ },
1091
+ {
1092
+ "epoch": 1.0581516095534786,
1093
+ "grad_norm": 0.45036646723747253,
1094
+ "learning_rate": 0.00010736964646747629,
1095
+ "loss": 0.4942,
1096
+ "step": 765
1097
+ },
1098
+ {
1099
+ "epoch": 1.0650744202146072,
1100
+ "grad_norm": 0.4810514748096466,
1101
+ "learning_rate": 0.00010616319025385089,
1102
+ "loss": 0.5105,
1103
+ "step": 770
1104
+ },
1105
+ {
1106
+ "epoch": 1.0719972308757355,
1107
+ "grad_norm": 0.43872886896133423,
1108
+ "learning_rate": 0.00010495583283908177,
1109
+ "loss": 0.5165,
1110
+ "step": 775
1111
+ },
1112
+ {
1113
+ "epoch": 1.078920041536864,
1114
+ "grad_norm": 0.40922561287879944,
1115
+ "learning_rate": 0.00010374775076678332,
1116
+ "loss": 0.5057,
1117
+ "step": 780
1118
+ },
1119
+ {
1120
+ "epoch": 1.0858428521979924,
1121
+ "grad_norm": 0.4093395471572876,
1122
+ "learning_rate": 0.00010253912068653146,
1123
+ "loss": 0.5153,
1124
+ "step": 785
1125
+ },
1126
+ {
1127
+ "epoch": 1.092765662859121,
1128
+ "grad_norm": 0.41029706597328186,
1129
+ "learning_rate": 0.00010133011932803378,
1130
+ "loss": 0.5204,
1131
+ "step": 790
1132
+ },
1133
+ {
1134
+ "epoch": 1.0996884735202492,
1135
+ "grad_norm": 0.4681115746498108,
1136
+ "learning_rate": 0.00010012092347528705,
1137
+ "loss": 0.5245,
1138
+ "step": 795
1139
+ },
1140
+ {
1141
+ "epoch": 1.1066112841813776,
1142
+ "grad_norm": 0.47305774688720703,
1143
+ "learning_rate": 9.891170994072793e-05,
1144
+ "loss": 0.5061,
1145
+ "step": 800
1146
+ },
1147
+ {
1148
+ "epoch": 1.113534094842506,
1149
+ "grad_norm": 0.44231101870536804,
1150
+ "learning_rate": 9.770265553937831e-05,
1151
+ "loss": 0.5092,
1152
+ "step": 805
1153
+ },
1154
+ {
1155
+ "epoch": 1.1204569055036344,
1156
+ "grad_norm": 0.4433559775352478,
1157
+ "learning_rate": 9.649393706299127e-05,
1158
+ "loss": 0.509,
1159
+ "step": 810
1160
  },
1161
  {
1162
+ "epoch": 1.127379716164763,
1163
+ "grad_norm": 0.45105186104774475,
1164
+ "learning_rate": 9.52857312541998e-05,
1165
+ "loss": 0.4962,
1166
+ "step": 815
1167
+ },
1168
+ {
1169
+ "epoch": 1.1343025268258913,
1170
+ "grad_norm": 0.40736016631126404,
1171
+ "learning_rate": 9.407821478067304e-05,
1172
+ "loss": 0.4962,
1173
+ "step": 820
1174
+ },
1175
+ {
1176
+ "epoch": 1.1412253374870198,
1177
+ "grad_norm": 0.4399997889995575,
1178
+ "learning_rate": 9.287156420928342e-05,
1179
+ "loss": 0.4985,
1180
+ "step": 825
1181
+ },
1182
+ {
1183
+ "epoch": 1.1481481481481481,
1184
+ "grad_norm": 0.4402114152908325,
1185
+ "learning_rate": 9.166595598028832e-05,
1186
+ "loss": 0.5181,
1187
+ "step": 830
1188
+ },
1189
+ {
1190
+ "epoch": 1.1550709588092767,
1191
+ "grad_norm": 0.4472050666809082,
1192
+ "learning_rate": 9.046156638153056e-05,
1193
+ "loss": 0.4966,
1194
+ "step": 835
1195
+ },
1196
+ {
1197
+ "epoch": 1.161993769470405,
1198
+ "grad_norm": 0.4352630376815796,
1199
+ "learning_rate": 8.925857152266103e-05,
1200
+ "loss": 0.4916,
1201
+ "step": 840
1202
+ },
1203
+ {
1204
+ "epoch": 1.1689165801315333,
1205
+ "grad_norm": 0.47437822818756104,
1206
+ "learning_rate": 8.805714730938728e-05,
1207
+ "loss": 0.5003,
1208
+ "step": 845
1209
+ },
1210
+ {
1211
+ "epoch": 1.1758393907926619,
1212
+ "grad_norm": 0.39329445362091064,
1213
+ "learning_rate": 8.685746941775219e-05,
1214
+ "loss": 0.497,
1215
+ "step": 850
1216
+ },
1217
+ {
1218
+ "epoch": 1.1827622014537902,
1219
+ "grad_norm": 0.4390949606895447,
1220
+ "learning_rate": 8.565971326844584e-05,
1221
+ "loss": 0.5101,
1222
+ "step": 855
1223
+ },
1224
+ {
1225
+ "epoch": 1.1896850121149187,
1226
+ "grad_norm": 0.421446293592453,
1227
+ "learning_rate": 8.44640540011553e-05,
1228
+ "loss": 0.5153,
1229
+ "step": 860
1230
+ },
1231
+ {
1232
+ "epoch": 1.196607822776047,
1233
+ "grad_norm": 0.4009932577610016,
1234
+ "learning_rate": 8.327066644895469e-05,
1235
+ "loss": 0.4998,
1236
+ "step": 865
1237
+ },
1238
+ {
1239
+ "epoch": 1.2035306334371756,
1240
+ "grad_norm": 0.45494183897972107,
1241
+ "learning_rate": 8.207972511274095e-05,
1242
+ "loss": 0.5151,
1243
+ "step": 870
1244
+ },
1245
+ {
1246
+ "epoch": 1.210453444098304,
1247
+ "grad_norm": 0.43307504057884216,
1248
+ "learning_rate": 8.089140413571747e-05,
1249
+ "loss": 0.5038,
1250
+ "step": 875
1251
+ },
1252
+ {
1253
+ "epoch": 1.2173762547594325,
1254
+ "grad_norm": 0.4515590965747833,
1255
+ "learning_rate": 7.970587727793051e-05,
1256
+ "loss": 0.4888,
1257
+ "step": 880
1258
+ },
1259
+ {
1260
+ "epoch": 1.2242990654205608,
1261
+ "grad_norm": 0.44130006432533264,
1262
+ "learning_rate": 7.852331789086135e-05,
1263
+ "loss": 0.4943,
1264
+ "step": 885
1265
+ },
1266
+ {
1267
+ "epoch": 1.231221876081689,
1268
+ "grad_norm": 0.40823453664779663,
1269
+ "learning_rate": 7.73438988920784e-05,
1270
+ "loss": 0.5018,
1271
+ "step": 890
1272
+ },
1273
+ {
1274
+ "epoch": 1.2381446867428176,
1275
+ "grad_norm": 0.45329058170318604,
1276
+ "learning_rate": 7.616779273995252e-05,
1277
+ "loss": 0.5091,
1278
+ "step": 895
1279
+ },
1280
+ {
1281
+ "epoch": 1.245067497403946,
1282
+ "grad_norm": 0.4063715636730194,
1283
+ "learning_rate": 7.499517140843968e-05,
1284
+ "loss": 0.5069,
1285
+ "step": 900
1286
+ },
1287
+ {
1288
+ "epoch": 1.2519903080650745,
1289
+ "grad_norm": 0.3937000036239624,
1290
+ "learning_rate": 7.382620636193438e-05,
1291
+ "loss": 0.4861,
1292
+ "step": 905
1293
+ },
1294
+ {
1295
+ "epoch": 1.2589131187262028,
1296
+ "grad_norm": 0.4193803071975708,
1297
+ "learning_rate": 7.266106853019759e-05,
1298
+ "loss": 0.4946,
1299
+ "step": 910
1300
+ },
1301
+ {
1302
+ "epoch": 1.2658359293873311,
1303
+ "grad_norm": 0.42721274495124817,
1304
+ "learning_rate": 7.149992828336274e-05,
1305
+ "loss": 0.5046,
1306
+ "step": 915
1307
+ },
1308
+ {
1309
+ "epoch": 1.2727587400484597,
1310
+ "grad_norm": 0.400716632604599,
1311
+ "learning_rate": 7.034295540702397e-05,
1312
+ "loss": 0.4893,
1313
+ "step": 920
1314
+ },
1315
+ {
1316
+ "epoch": 1.2796815507095882,
1317
+ "grad_norm": 0.41494622826576233,
1318
+ "learning_rate": 6.919031907740922e-05,
1319
+ "loss": 0.5035,
1320
+ "step": 925
1321
+ },
1322
+ {
1323
+ "epoch": 1.2866043613707165,
1324
+ "grad_norm": 0.4320232570171356,
1325
+ "learning_rate": 6.804218783664288e-05,
1326
+ "loss": 0.5042,
1327
+ "step": 930
1328
+ },
1329
+ {
1330
+ "epoch": 1.2935271720318449,
1331
+ "grad_norm": 0.4132310748100281,
1332
+ "learning_rate": 6.689872956810102e-05,
1333
+ "loss": 0.5009,
1334
+ "step": 935
1335
+ },
1336
+ {
1337
+ "epoch": 1.3004499826929734,
1338
+ "grad_norm": 0.4009295403957367,
1339
+ "learning_rate": 6.5760111471863e-05,
1340
+ "loss": 0.4961,
1341
+ "step": 940
1342
+ },
1343
+ {
1344
+ "epoch": 1.3073727933541017,
1345
+ "grad_norm": 0.40771910548210144,
1346
+ "learning_rate": 6.46265000402628e-05,
1347
+ "loss": 0.5076,
1348
+ "step": 945
1349
+ },
1350
+ {
1351
+ "epoch": 1.3142956040152303,
1352
+ "grad_norm": 0.41762447357177734,
1353
+ "learning_rate": 6.349806103354417e-05,
1354
+ "loss": 0.495,
1355
+ "step": 950
1356
+ },
1357
+ {
1358
+ "epoch": 1.3212184146763586,
1359
+ "grad_norm": 0.4337986707687378,
1360
+ "learning_rate": 6.23749594556226e-05,
1361
+ "loss": 0.4887,
1362
+ "step": 955
1363
+ },
1364
+ {
1365
+ "epoch": 1.328141225337487,
1366
+ "grad_norm": 0.4177555441856384,
1367
+ "learning_rate": 6.125735952995775e-05,
1368
+ "loss": 0.4824,
1369
+ "step": 960
1370
+ },
1371
+ {
1372
+ "epoch": 1.3350640359986155,
1373
+ "grad_norm": 0.4223066568374634,
1374
+ "learning_rate": 6.0145424675540394e-05,
1375
+ "loss": 0.4928,
1376
+ "step": 965
1377
+ },
1378
+ {
1379
+ "epoch": 1.341986846659744,
1380
+ "grad_norm": 0.406967431306839,
1381
+ "learning_rate": 5.9039317482996735e-05,
1382
+ "loss": 0.4996,
1383
+ "step": 970
1384
+ },
1385
+ {
1386
+ "epoch": 1.3489096573208723,
1387
+ "grad_norm": 0.4023487865924835,
1388
+ "learning_rate": 5.793919969081374e-05,
1389
+ "loss": 0.4833,
1390
+ "step": 975
1391
+ },
1392
+ {
1393
+ "epoch": 1.3558324679820006,
1394
+ "grad_norm": 0.38443300127983093,
1395
+ "learning_rate": 5.68452321616894e-05,
1396
+ "loss": 0.4741,
1397
+ "step": 980
1398
+ },
1399
+ {
1400
+ "epoch": 1.3627552786431292,
1401
+ "grad_norm": 0.3877335786819458,
1402
+ "learning_rate": 5.5757574859010886e-05,
1403
+ "loss": 0.4897,
1404
+ "step": 985
1405
+ },
1406
+ {
1407
+ "epoch": 1.3696780893042575,
1408
+ "grad_norm": 0.3927278518676758,
1409
+ "learning_rate": 5.467638682346403e-05,
1410
+ "loss": 0.4733,
1411
+ "step": 990
1412
+ },
1413
+ {
1414
+ "epoch": 1.376600899965386,
1415
+ "grad_norm": 0.42467862367630005,
1416
+ "learning_rate": 5.3601826149777966e-05,
1417
+ "loss": 0.4903,
1418
+ "step": 995
1419
+ },
1420
+ {
1421
+ "epoch": 1.3835237106265144,
1422
+ "grad_norm": 0.42438316345214844,
1423
+ "learning_rate": 5.25340499636082e-05,
1424
+ "loss": 0.4818,
1425
+ "step": 1000
1426
+ },
1427
+ {
1428
+ "epoch": 1.3904465212876427,
1429
+ "grad_norm": 0.408420592546463,
1430
+ "learning_rate": 5.147321439856091e-05,
1431
+ "loss": 0.4879,
1432
+ "step": 1005
1433
+ },
1434
+ {
1435
+ "epoch": 1.3973693319487712,
1436
+ "grad_norm": 0.4005957543849945,
1437
+ "learning_rate": 5.041947457336274e-05,
1438
+ "loss": 0.5015,
1439
+ "step": 1010
1440
+ },
1441
+ {
1442
+ "epoch": 1.4042921426098995,
1443
+ "grad_norm": 0.38740673661231995,
1444
+ "learning_rate": 4.937298456917889e-05,
1445
+ "loss": 0.4902,
1446
+ "step": 1015
1447
+ },
1448
+ {
1449
+ "epoch": 1.411214953271028,
1450
+ "grad_norm": 0.4314693510532379,
1451
+ "learning_rate": 4.8333897407082896e-05,
1452
+ "loss": 0.4825,
1453
+ "step": 1020
1454
+ },
1455
+ {
1456
+ "epoch": 1.4181377639321564,
1457
+ "grad_norm": 0.37632128596305847,
1458
+ "learning_rate": 4.7302365025681206e-05,
1459
+ "loss": 0.4879,
1460
+ "step": 1025
1461
+ },
1462
+ {
1463
+ "epoch": 1.425060574593285,
1464
+ "grad_norm": 0.4002823531627655,
1465
+ "learning_rate": 4.627853825889642e-05,
1466
+ "loss": 0.4849,
1467
+ "step": 1030
1468
+ },
1469
+ {
1470
+ "epoch": 1.4319833852544133,
1471
+ "grad_norm": 0.42868149280548096,
1472
+ "learning_rate": 4.526256681391192e-05,
1473
+ "loss": 0.4774,
1474
+ "step": 1035
1475
+ },
1476
+ {
1477
+ "epoch": 1.4389061959155418,
1478
+ "grad_norm": 0.4506429135799408,
1479
+ "learning_rate": 4.4254599249281016e-05,
1480
+ "loss": 0.4634,
1481
+ "step": 1040
1482
+ },
1483
+ {
1484
+ "epoch": 1.4458290065766701,
1485
+ "grad_norm": 0.40321943163871765,
1486
+ "learning_rate": 4.3254782953204375e-05,
1487
+ "loss": 0.4816,
1488
+ "step": 1045
1489
+ },
1490
+ {
1491
+ "epoch": 1.4527518172377984,
1492
+ "grad_norm": 0.4356391429901123,
1493
+ "learning_rate": 4.2263264121978505e-05,
1494
+ "loss": 0.491,
1495
+ "step": 1050
1496
+ },
1497
+ {
1498
+ "epoch": 1.459674627898927,
1499
+ "grad_norm": 0.3813280165195465,
1500
+ "learning_rate": 4.12801877386183e-05,
1501
+ "loss": 0.4698,
1502
+ "step": 1055
1503
+ },
1504
+ {
1505
+ "epoch": 1.4665974385600553,
1506
+ "grad_norm": 0.41611161828041077,
1507
+ "learning_rate": 4.030569755165736e-05,
1508
+ "loss": 0.4833,
1509
+ "step": 1060
1510
+ },
1511
+ {
1512
+ "epoch": 1.4735202492211839,
1513
+ "grad_norm": 0.40740102529525757,
1514
+ "learning_rate": 3.9339936054128466e-05,
1515
+ "loss": 0.4703,
1516
+ "step": 1065
1517
+ },
1518
+ {
1519
+ "epoch": 1.4804430598823122,
1520
+ "grad_norm": 0.3984625041484833,
1521
+ "learning_rate": 3.8383044462727826e-05,
1522
+ "loss": 0.4628,
1523
+ "step": 1070
1524
+ },
1525
+ {
1526
+ "epoch": 1.4873658705434407,
1527
+ "grad_norm": 0.42874425649642944,
1528
+ "learning_rate": 3.7435162697166107e-05,
1529
+ "loss": 0.4801,
1530
+ "step": 1075
1531
+ },
1532
+ {
1533
+ "epoch": 1.494288681204569,
1534
+ "grad_norm": 0.3962753117084503,
1535
+ "learning_rate": 3.649642935970859e-05,
1536
+ "loss": 0.4805,
1537
+ "step": 1080
1538
+ },
1539
+ {
1540
+ "epoch": 1.5012114918656976,
1541
+ "grad_norm": 0.37871405482292175,
1542
+ "learning_rate": 3.556698171490871e-05,
1543
+ "loss": 0.4718,
1544
+ "step": 1085
1545
+ },
1546
+ {
1547
+ "epoch": 1.508134302526826,
1548
+ "grad_norm": 0.39410313963890076,
1549
+ "learning_rate": 3.464695566953644e-05,
1550
+ "loss": 0.4845,
1551
+ "step": 1090
1552
+ },
1553
+ {
1554
+ "epoch": 1.5150571131879542,
1555
+ "grad_norm": 0.4159853458404541,
1556
+ "learning_rate": 3.373648575270576e-05,
1557
+ "loss": 0.4823,
1558
+ "step": 1095
1559
+ },
1560
+ {
1561
+ "epoch": 1.5219799238490828,
1562
+ "grad_norm": 0.4162219166755676,
1563
+ "learning_rate": 3.283570509620344e-05,
1564
+ "loss": 0.4819,
1565
+ "step": 1100
1566
+ },
1567
+ {
1568
+ "epoch": 1.5289027345102113,
1569
+ "grad_norm": 0.39965444803237915,
1570
+ "learning_rate": 3.1944745415021915e-05,
1571
+ "loss": 0.4715,
1572
+ "step": 1105
1573
+ },
1574
+ {
1575
+ "epoch": 1.5358255451713396,
1576
+ "grad_norm": 0.4039742052555084,
1577
+ "learning_rate": 3.106373698809958e-05,
1578
+ "loss": 0.4673,
1579
+ "step": 1110
1580
+ },
1581
+ {
1582
+ "epoch": 1.542748355832468,
1583
+ "grad_norm": 0.3849252164363861,
1584
+ "learning_rate": 3.0192808639271065e-05,
1585
+ "loss": 0.4758,
1586
+ "step": 1115
1587
+ },
1588
+ {
1589
+ "epoch": 1.5496711664935963,
1590
+ "grad_norm": 0.36080726981163025,
1591
+ "learning_rate": 2.933208771843018e-05,
1592
+ "loss": 0.4818,
1593
+ "step": 1120
1594
+ },
1595
+ {
1596
+ "epoch": 1.5565939771547248,
1597
+ "grad_norm": 0.3905937373638153,
1598
+ "learning_rate": 2.8481700082908268e-05,
1599
+ "loss": 0.4668,
1600
+ "step": 1125
1601
+ },
1602
+ {
1603
+ "epoch": 1.5635167878158533,
1604
+ "grad_norm": 0.3504320979118347,
1605
+ "learning_rate": 2.7641770079071127e-05,
1606
+ "loss": 0.4648,
1607
+ "step": 1130
1608
+ },
1609
+ {
1610
+ "epoch": 1.5704395984769817,
1611
+ "grad_norm": 0.4071274697780609,
1612
+ "learning_rate": 2.681242052413656e-05,
1613
+ "loss": 0.4644,
1614
+ "step": 1135
1615
+ },
1616
+ {
1617
+ "epoch": 1.57736240913811,
1618
+ "grad_norm": 0.388864666223526,
1619
+ "learning_rate": 2.59937726882158e-05,
1620
+ "loss": 0.4689,
1621
+ "step": 1140
1622
+ },
1623
+ {
1624
+ "epoch": 1.5842852197992385,
1625
+ "grad_norm": 0.4169630706310272,
1626
+ "learning_rate": 2.5185946276580918e-05,
1627
+ "loss": 0.4686,
1628
+ "step": 1145
1629
+ },
1630
+ {
1631
+ "epoch": 1.591208030460367,
1632
+ "grad_norm": 0.3965269923210144,
1633
+ "learning_rate": 2.4389059412161087e-05,
1634
+ "loss": 0.4676,
1635
+ "step": 1150
1636
+ },
1637
+ {
1638
+ "epoch": 1.5981308411214954,
1639
+ "grad_norm": 0.38269945979118347,
1640
+ "learning_rate": 2.3603228618270356e-05,
1641
+ "loss": 0.4723,
1642
+ "step": 1155
1643
+ },
1644
+ {
1645
+ "epoch": 1.6050536517826237,
1646
+ "grad_norm": 0.4088960587978363,
1647
+ "learning_rate": 2.2828568801569283e-05,
1648
+ "loss": 0.454,
1649
+ "step": 1160
1650
+ },
1651
+ {
1652
+ "epoch": 1.611976462443752,
1653
+ "grad_norm": 0.40200474858283997,
1654
+ "learning_rate": 2.2065193235262782e-05,
1655
+ "loss": 0.4675,
1656
+ "step": 1165
1657
+ },
1658
+ {
1659
+ "epoch": 1.6188992731048806,
1660
+ "grad_norm": 0.4130989909172058,
1661
+ "learning_rate": 2.1313213542537058e-05,
1662
+ "loss": 0.46,
1663
+ "step": 1170
1664
+ },
1665
+ {
1666
+ "epoch": 1.6258220837660091,
1667
+ "grad_norm": 0.42108625173568726,
1668
+ "learning_rate": 2.0572739680237717e-05,
1669
+ "loss": 0.4763,
1670
+ "step": 1175
1671
+ },
1672
+ {
1673
+ "epoch": 1.6327448944271374,
1674
+ "grad_norm": 0.38774868845939636,
1675
+ "learning_rate": 1.984387992279153e-05,
1676
+ "loss": 0.4735,
1677
+ "step": 1180
1678
+ },
1679
+ {
1680
+ "epoch": 1.6396677050882658,
1681
+ "grad_norm": 0.39550262689590454,
1682
+ "learning_rate": 1.912674084637409e-05,
1683
+ "loss": 0.4572,
1684
+ "step": 1185
1685
+ },
1686
+ {
1687
+ "epoch": 1.6465905157493943,
1688
+ "grad_norm": 0.3853910267353058,
1689
+ "learning_rate": 1.8421427313326046e-05,
1690
+ "loss": 0.4677,
1691
+ "step": 1190
1692
+ },
1693
+ {
1694
+ "epoch": 1.6535133264105226,
1695
+ "grad_norm": 0.4146302342414856,
1696
+ "learning_rate": 1.7728042456819704e-05,
1697
+ "loss": 0.4637,
1698
+ "step": 1195
1699
+ },
1700
+ {
1701
+ "epoch": 1.6604361370716512,
1702
+ "grad_norm": 0.39156144857406616,
1703
+ "learning_rate": 1.7046687665778717e-05,
1704
+ "loss": 0.4601,
1705
+ "step": 1200
1706
+ },
1707
+ {
1708
+ "epoch": 1.6673589477327795,
1709
+ "grad_norm": 0.3790566325187683,
1710
+ "learning_rate": 1.6377462570052438e-05,
1711
+ "loss": 0.4594,
1712
+ "step": 1205
1713
+ },
1714
+ {
1715
+ "epoch": 1.6742817583939078,
1716
+ "grad_norm": 0.38195911049842834,
1717
+ "learning_rate": 1.5720465025847987e-05,
1718
+ "loss": 0.4644,
1719
+ "step": 1210
1720
+ },
1721
+ {
1722
+ "epoch": 1.6812045690550363,
1723
+ "grad_norm": 0.391541063785553,
1724
+ "learning_rate": 1.5075791101421166e-05,
1725
+ "loss": 0.4617,
1726
+ "step": 1215
1727
+ },
1728
+ {
1729
+ "epoch": 1.6881273797161649,
1730
+ "grad_norm": 0.391732394695282,
1731
+ "learning_rate": 1.4443535063029279e-05,
1732
+ "loss": 0.4585,
1733
+ "step": 1220
1734
+ },
1735
+ {
1736
+ "epoch": 1.6950501903772932,
1737
+ "grad_norm": 0.4002660810947418,
1738
+ "learning_rate": 1.3823789361147044e-05,
1739
+ "loss": 0.4604,
1740
+ "step": 1225
1741
+ },
1742
+ {
1743
+ "epoch": 1.7019730010384215,
1744
+ "grad_norm": 0.3701544404029846,
1745
+ "learning_rate": 1.3216644616948282e-05,
1746
+ "loss": 0.4595,
1747
+ "step": 1230
1748
+ },
1749
+ {
1750
+ "epoch": 1.70889581169955,
1751
+ "grad_norm": 0.36977705359458923,
1752
+ "learning_rate": 1.2622189609054979e-05,
1753
+ "loss": 0.4575,
1754
+ "step": 1235
1755
+ },
1756
+ {
1757
+ "epoch": 1.7158186223606784,
1758
+ "grad_norm": 0.4012480080127716,
1759
+ "learning_rate": 1.2040511260555831e-05,
1760
+ "loss": 0.4491,
1761
+ "step": 1240
1762
+ },
1763
+ {
1764
+ "epoch": 1.722741433021807,
1765
+ "grad_norm": 0.3966807425022125,
1766
+ "learning_rate": 1.1471694626296049e-05,
1767
+ "loss": 0.457,
1768
+ "step": 1245
1769
+ },
1770
+ {
1771
+ "epoch": 1.7296642436829353,
1772
+ "grad_norm": 0.39245152473449707,
1773
+ "learning_rate": 1.0915822880440308e-05,
1774
+ "loss": 0.4484,
1775
+ "step": 1250
1776
+ },
1777
+ {
1778
+ "epoch": 1.7365870543440636,
1779
+ "grad_norm": 0.38238054513931274,
1780
+ "learning_rate": 1.037297730431085e-05,
1781
+ "loss": 0.4391,
1782
+ "step": 1255
1783
+ },
1784
+ {
1785
+ "epoch": 1.7435098650051921,
1786
+ "grad_norm": 0.38661518692970276,
1787
+ "learning_rate": 9.843237274502282e-06,
1788
+ "loss": 0.4701,
1789
+ "step": 1260
1790
+ },
1791
+ {
1792
+ "epoch": 1.7504326756663207,
1793
+ "grad_norm": 0.3944835066795349,
1794
+ "learning_rate": 9.326680251274777e-06,
1795
+ "loss": 0.448,
1796
+ "step": 1265
1797
+ },
1798
+ {
1799
+ "epoch": 1.757355486327449,
1800
+ "grad_norm": 0.3993683457374573,
1801
+ "learning_rate": 8.823381767227667e-06,
1802
+ "loss": 0.4577,
1803
+ "step": 1270
1804
+ },
1805
+ {
1806
+ "epoch": 1.7642782969885773,
1807
+ "grad_norm": 0.385385125875473,
1808
+ "learning_rate": 8.333415416254831e-06,
1809
+ "loss": 0.4601,
1810
+ "step": 1275
1811
+ },
1812
+ {
1813
+ "epoch": 1.7712011076497056,
1814
+ "grad_norm": 0.37722861766815186,
1815
+ "learning_rate": 7.856852842783547e-06,
1816
+ "loss": 0.4625,
1817
+ "step": 1280
1818
+ },
1819
+ {
1820
+ "epoch": 1.7781239183108342,
1821
+ "grad_norm": 0.37356412410736084,
1822
+ "learning_rate": 7.393763731298387e-06,
1823
+ "loss": 0.4703,
1824
+ "step": 1285
1825
+ },
1826
+ {
1827
+ "epoch": 1.7850467289719627,
1828
+ "grad_norm": 0.3780430555343628,
1829
+ "learning_rate": 6.944215796151765e-06,
1830
+ "loss": 0.4515,
1831
+ "step": 1290
1832
+ },
1833
+ {
1834
+ "epoch": 1.791969539633091,
1835
+ "grad_norm": 0.40583980083465576,
1836
+ "learning_rate": 6.5082747716625255e-06,
1837
+ "loss": 0.4462,
1838
+ "step": 1295
1839
+ },
1840
+ {
1841
+ "epoch": 1.7988923502942193,
1842
+ "grad_norm": 0.3829346001148224,
1843
+ "learning_rate": 6.086004402504098e-06,
1844
+ "loss": 0.4605,
1845
+ "step": 1300
1846
+ },
1847
+ {
1848
+ "epoch": 1.8058151609553479,
1849
+ "grad_norm": 0.3809064030647278,
1850
+ "learning_rate": 5.6774664343834495e-06,
1851
+ "loss": 0.4607,
1852
+ "step": 1305
1853
+ },
1854
+ {
1855
+ "epoch": 1.8127379716164764,
1856
+ "grad_norm": 0.38634175062179565,
1857
+ "learning_rate": 5.2827206050125585e-06,
1858
+ "loss": 0.4671,
1859
+ "step": 1310
1860
+ },
1861
+ {
1862
+ "epoch": 1.8196607822776047,
1863
+ "grad_norm": 0.39494749903678894,
1864
+ "learning_rate": 4.9018246353732755e-06,
1865
+ "loss": 0.4684,
1866
+ "step": 1315
1867
+ },
1868
+ {
1869
+ "epoch": 1.826583592938733,
1870
+ "grad_norm": 0.3794882595539093,
1871
+ "learning_rate": 4.534834221277262e-06,
1872
+ "loss": 0.4527,
1873
+ "step": 1320
1874
+ },
1875
+ {
1876
+ "epoch": 1.8335064035998614,
1877
+ "grad_norm": 0.37009429931640625,
1878
+ "learning_rate": 4.181803025221898e-06,
1879
+ "loss": 0.4614,
1880
+ "step": 1325
1881
+ },
1882
+ {
1883
+ "epoch": 1.84042921426099,
1884
+ "grad_norm": 0.3789846897125244,
1885
+ "learning_rate": 3.842782668543599e-06,
1886
+ "loss": 0.4512,
1887
+ "step": 1330
1888
+ },
1889
+ {
1890
+ "epoch": 1.8473520249221185,
1891
+ "grad_norm": 0.33748793601989746,
1892
+ "learning_rate": 3.5178227238696347e-06,
1893
+ "loss": 0.4673,
1894
+ "step": 1335
1895
+ },
1896
+ {
1897
+ "epoch": 1.8542748355832468,
1898
+ "grad_norm": 0.3598109483718872,
1899
+ "learning_rate": 3.2069707078694057e-06,
1900
+ "loss": 0.4459,
1901
+ "step": 1340
1902
+ },
1903
+ {
1904
+ "epoch": 1.8611976462443751,
1905
+ "grad_norm": 0.35582590103149414,
1906
+ "learning_rate": 2.9102720743064527e-06,
1907
+ "loss": 0.447,
1908
+ "step": 1345
1909
+ },
1910
+ {
1911
+ "epoch": 1.8681204569055037,
1912
+ "grad_norm": 0.3721593916416168,
1913
+ "learning_rate": 2.6277702073919997e-06,
1914
+ "loss": 0.4498,
1915
+ "step": 1350
1916
+ },
1917
+ {
1918
+ "epoch": 1.8750432675666322,
1919
+ "grad_norm": 0.3712756931781769,
1920
+ "learning_rate": 2.3595064154412374e-06,
1921
+ "loss": 0.456,
1922
+ "step": 1355
1923
+ },
1924
+ {
1925
+ "epoch": 1.8819660782277605,
1926
+ "grad_norm": 0.37412285804748535,
1927
+ "learning_rate": 2.1055199248330546e-06,
1928
+ "loss": 0.4454,
1929
+ "step": 1360
1930
+ },
1931
+ {
1932
+ "epoch": 1.8888888888888888,
1933
+ "grad_norm": 0.3859015107154846,
1934
+ "learning_rate": 1.8658478742742625e-06,
1935
+ "loss": 0.4529,
1936
+ "step": 1365
1937
+ },
1938
+ {
1939
+ "epoch": 1.8958116995500172,
1940
+ "grad_norm": 0.37355226278305054,
1941
+ "learning_rate": 1.6405253093690343e-06,
1942
+ "loss": 0.4565,
1943
+ "step": 1370
1944
+ },
1945
+ {
1946
+ "epoch": 1.9027345102111457,
1947
+ "grad_norm": 0.4014715552330017,
1948
+ "learning_rate": 1.429585177494419e-06,
1949
+ "loss": 0.4556,
1950
+ "step": 1375
1951
+ },
1952
+ {
1953
+ "epoch": 1.9096573208722742,
1954
+ "grad_norm": 0.40377333760261536,
1955
+ "learning_rate": 1.2330583229827053e-06,
1956
+ "loss": 0.4676,
1957
+ "step": 1380
1958
+ },
1959
+ {
1960
+ "epoch": 1.9165801315334026,
1961
+ "grad_norm": 0.3818195164203644,
1962
+ "learning_rate": 1.0509734826112394e-06,
1963
+ "loss": 0.4483,
1964
+ "step": 1385
1965
+ },
1966
+ {
1967
+ "epoch": 1.9235029421945309,
1968
+ "grad_norm": 0.40282756090164185,
1969
+ "learning_rate": 8.833572814004321e-07,
1970
+ "loss": 0.4402,
1971
+ "step": 1390
1972
+ },
1973
+ {
1974
+ "epoch": 1.9304257528556594,
1975
+ "grad_norm": 0.37681761384010315,
1976
+ "learning_rate": 7.302342287205943e-07,
1977
+ "loss": 0.4464,
1978
+ "step": 1395
1979
+ },
1980
+ {
1981
+ "epoch": 1.937348563516788,
1982
+ "grad_norm": 0.3510221838951111,
1983
+ "learning_rate": 5.916267147080934e-07,
1984
+ "loss": 0.4634,
1985
+ "step": 1400
1986
+ },
1987
+ {
1988
+ "epoch": 1.9442713741779163,
1989
+ "grad_norm": 0.40750935673713684,
1990
+ "learning_rate": 4.6755500699138253e-07,
1991
+ "loss": 0.4537,
1992
+ "step": 1405
1993
+ },
1994
+ {
1995
+ "epoch": 1.9511941848390446,
1996
+ "grad_norm": 0.3761488199234009,
1997
+ "learning_rate": 3.5803724772740523e-07,
1998
+ "loss": 0.4427,
1999
+ "step": 1410
2000
+ },
2001
+ {
2002
+ "epoch": 1.958116995500173,
2003
+ "grad_norm": 0.3658927381038666,
2004
+ "learning_rate": 2.630894509488058e-07,
2005
+ "loss": 0.4639,
2006
+ "step": 1415
2007
+ },
2008
+ {
2009
+ "epoch": 1.9650398061613015,
2010
+ "grad_norm": 0.3706067204475403,
2011
+ "learning_rate": 1.8272550022230272e-07,
2012
+ "loss": 0.4559,
2013
+ "step": 1420
2014
+ },
2015
+ {
2016
+ "epoch": 1.97196261682243,
2017
+ "grad_norm": 0.38704028725624084,
2018
+ "learning_rate": 1.1695714661856816e-07,
2019
+ "loss": 0.4602,
2020
+ "step": 1425
2021
+ },
2022
+ {
2023
+ "epoch": 1.9788854274835583,
2024
+ "grad_norm": 0.3871050179004669,
2025
+ "learning_rate": 6.579400699397998e-08,
2026
+ "loss": 0.4508,
2027
+ "step": 1430
2028
+ },
2029
+ {
2030
+ "epoch": 1.9858082381446867,
2031
+ "grad_norm": 0.3808486759662628,
2032
+ "learning_rate": 2.9243562584402305e-08,
2033
+ "loss": 0.4511,
2034
+ "step": 1435
2035
+ },
2036
+ {
2037
+ "epoch": 1.9927310488058152,
2038
+ "grad_norm": 0.3793506622314453,
2039
+ "learning_rate": 7.311157911249478e-09,
2040
+ "loss": 0.456,
2041
+ "step": 1440
2042
+ },
2043
+ {
2044
+ "epoch": 1.9982692973347178,
2045
+ "eval_loss": 0.5922103524208069,
2046
+ "eval_runtime": 79.546,
2047
+ "eval_samples_per_second": 26.01,
2048
+ "eval_steps_per_second": 1.634,
2049
+ "step": 1444
2050
+ },
2051
+ {
2052
+ "epoch": 1.9982692973347178,
2053
+ "step": 1444,
2054
+ "total_flos": 7.179552437307441e+18,
2055
+ "train_loss": 0.6496460468319974,
2056
+ "train_runtime": 14397.0712,
2057
+ "train_samples_per_second": 6.42,
2058
+ "train_steps_per_second": 0.1
2059
  }
2060
  ],
2061
  "logging_steps": 5,
2062
+ "max_steps": 1444,
2063
  "num_input_tokens_seen": 0,
2064
+ "num_train_epochs": 2,
2065
  "save_steps": 500,
2066
  "stateful_callbacks": {
2067
  "TrainerControl": {
 
2069
  "should_epoch_stop": false,
2070
  "should_evaluate": false,
2071
  "should_log": false,
2072
+ "should_save": false,
2073
+ "should_training_stop": false
2074
  },
2075
  "attributes": {}
2076
  }
2077
  },
2078
+ "total_flos": 7.179552437307441e+18,
2079
  "train_batch_size": 4,
2080
  "trial_name": null,
2081
  "trial_params": null
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:52e38047350b995275b0facde1c76bf4478da59cb980ea74aed2d82c48c6ffc6
3
- size 7032
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a11107f3932b3b98e2c8bacb84cc12ff21b6d598e4c9cf1a6dde9196e85c6ecd
3
+ size 7096