AlekseyKorshuk commited on
Commit
fa424a4
1 Parent(s): 90c90c8

huggingartists

Browse files
README.md CHANGED
@@ -45,15 +45,15 @@ from datasets import load_dataset
45
  dataset = load_dataset("huggingartists/6ix9ine")
46
  ```
47
 
48
- [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/1ubrxjqi/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
49
 
50
  ## Training procedure
51
 
52
  The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on 6ix9ine's lyrics.
53
 
54
- Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/x99noxul) for full transparency and reproducibility.
55
 
56
- At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/x99noxul/artifacts) is logged and versioned.
57
 
58
  ## How to use
59
 
45
  dataset = load_dataset("huggingartists/6ix9ine")
46
  ```
47
 
48
+ [Explore the data](https://wandb.ai/huggingartists/huggingartists/runs/33owzv4t/artifacts), which is tracked with [W&B artifacts](https://docs.wandb.com/artifacts) at every step of the pipeline.
49
 
50
  ## Training procedure
51
 
52
  The model is based on a pre-trained [GPT-2](https://huggingface.co/gpt2) which is fine-tuned on 6ix9ine's lyrics.
53
 
54
+ Hyperparameters and metrics are recorded in the [W&B training run](https://wandb.ai/huggingartists/huggingartists/runs/22q1oxft) for full transparency and reproducibility.
55
 
56
+ At the end of training, [the final model](https://wandb.ai/huggingartists/huggingartists/runs/22q1oxft/artifacts) is logged and versioned.
57
 
58
  ## How to use
59
 
config.json CHANGED
@@ -1,5 +1,5 @@
1
  {
2
- "_name_or_path": "gpt2",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
  "GPT2LMHeadModel"
1
  {
2
+ "_name_or_path": "huggingartists/6ix9ine",
3
  "activation_function": "gelu_new",
4
  "architectures": [
5
  "GPT2LMHeadModel"
evaluation.txt CHANGED
@@ -1 +1 @@
1
- {"eval_loss": 3.22310209274292, "eval_runtime": 2.9511, "eval_samples_per_second": 21.009, "eval_steps_per_second": 2.711, "epoch": 1.0}
1
+ {"eval_loss": 3.0470073223114014, "eval_runtime": 2.6704, "eval_samples_per_second": 20.596, "eval_steps_per_second": 2.621, "epoch": 2.0}
flax_model.msgpack CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:21a29eb776edb4b712524c6abba8c4860f899e3635fd68ff7cc59a3d49975c8a
3
  size 497764120
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8df15ff3bf9d8a1748a052761e63c3b2f0e766961f7279a8de10378dfaeaf585
3
  size 497764120
optimizer.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:635c24ec0c45101d2cc83a7e5762597d78a3de8b7f6009164decc1d7593a427c
3
  size 995603825
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fb7c973b07fc7925a7773a7a9b8e96db55101b49d6bf109cfa76b63b9f7b450e
3
  size 995603825
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d406ad882369a3e603cd147a311f51ecf1dc400c235f725ce5f0ad6a38a17306
3
  size 510403817
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fd1cf3ab818bb31d83b57f1db70e0d084389c521eecdf3cbb96f980dd4a48168
3
  size 510403817
rng_state.pth CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:803cf259ab295010aaedc7ed0ba89a9d393a8018b1e858e6b80cb18034f8b057
3
  size 14503
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:29d7a8e82d5104f8855899362d2c29661c48c9cc84dceb8ac5bbc637db2337f5
3
  size 14503
scheduler.pt CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8d4369506dedd54e65c6967a3362585e3aa7b5afb87732c009d3c12491c2fbd1
3
  size 623
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00dd6b6c01d14ec8bb955c14a18bd11e818291c6d3cf3ac44048cedcc0186d76
3
  size 623
tokenizer_config.json CHANGED
@@ -1 +1 @@
1
- {"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2", "tokenizer_class": "GPT2Tokenizer"}
1
+ {"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "huggingartists/6ix9ine", "tokenizer_class": "GPT2Tokenizer"}
trainer_state.json CHANGED
@@ -1,72 +1,128 @@
1
  {
2
- "best_metric": 3.22310209274292,
3
- "best_model_checkpoint": "output/6ix9ine/checkpoint-41",
4
- "epoch": 1.0,
5
- "global_step": 41,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
  "epoch": 0.12,
12
- "learning_rate": 0.0001322266681456037,
13
- "loss": 3.9471,
14
  "step": 5
15
  },
16
  {
17
  "epoch": 0.24,
18
- "learning_rate": 0.00011802778132101399,
19
- "loss": 3.8762,
20
  "step": 10
21
  },
22
  {
23
- "epoch": 0.37,
24
- "learning_rate": 9.666210850995393e-05,
25
- "loss": 3.5952,
26
  "step": 15
27
  },
28
  {
29
- "epoch": 0.49,
30
- "learning_rate": 7.122756753113643e-05,
31
- "loss": 3.596,
32
  "step": 20
33
  },
34
  {
35
- "epoch": 0.61,
36
- "learning_rate": 4.541204214117672e-05,
37
- "loss": 3.577,
38
  "step": 25
39
  },
40
  {
41
- "epoch": 0.73,
42
- "learning_rate": 2.2958656968642224e-05,
43
- "loss": 3.469,
44
  "step": 30
45
  },
46
  {
47
- "epoch": 0.85,
48
- "learning_rate": 7.123042792471594e-06,
49
- "loss": 3.3983,
50
  "step": 35
51
  },
52
  {
53
- "epoch": 0.98,
54
- "learning_rate": 2.0128603879540573e-07,
55
- "loss": 3.4279,
56
  "step": 40
57
  },
58
  {
59
  "epoch": 1.0,
60
- "eval_loss": 3.22310209274292,
61
- "eval_runtime": 2.9378,
62
- "eval_samples_per_second": 21.104,
63
- "eval_steps_per_second": 2.723,
64
- "step": 41
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
65
  }
66
  ],
67
- "max_steps": 41,
68
- "num_train_epochs": 1,
69
- "total_flos": 42198663168000.0,
70
  "trial_name": null,
71
  "trial_params": null
72
  }
1
  {
2
+ "best_metric": 3.0470073223114014,
3
+ "best_model_checkpoint": "output/6ix9ine/checkpoint-84",
4
+ "epoch": 2.0,
5
+ "global_step": 84,
6
  "is_hyper_param_search": false,
7
  "is_local_process_zero": true,
8
  "is_world_process_zero": true,
9
  "log_history": [
10
  {
11
  "epoch": 0.12,
12
+ "learning_rate": 4.742060843007589e-06,
13
+ "loss": 3.316,
14
  "step": 5
15
  },
16
  {
17
  "epoch": 0.24,
18
+ "learning_rate": 1.8312641592473922e-05,
19
+ "loss": 3.5414,
20
  "step": 10
21
  },
22
  {
23
+ "epoch": 0.36,
24
+ "learning_rate": 3.88355754965355e-05,
25
+ "loss": 3.3994,
26
  "step": 15
27
  },
28
  {
29
+ "epoch": 0.48,
30
+ "learning_rate": 6.347351557997133e-05,
31
+ "loss": 3.2221,
32
  "step": 20
33
  },
34
  {
35
+ "epoch": 0.6,
36
+ "learning_rate": 8.882020496458803e-05,
37
+ "loss": 3.2784,
38
  "step": 25
39
  },
40
  {
41
+ "epoch": 0.71,
42
+ "learning_rate": 0.00011137140040750911,
43
+ "loss": 3.376,
44
  "step": 30
45
  },
46
  {
47
+ "epoch": 0.83,
48
+ "learning_rate": 0.00012800934269961248,
49
+ "loss": 3.0065,
50
  "step": 35
51
  },
52
  {
53
+ "epoch": 0.95,
54
+ "learning_rate": 0.0001364337946790438,
55
+ "loss": 3.118,
56
  "step": 40
57
  },
58
  {
59
  "epoch": 1.0,
60
+ "eval_loss": 3.159865617752075,
61
+ "eval_runtime": 2.6213,
62
+ "eval_samples_per_second": 20.982,
63
+ "eval_steps_per_second": 2.67,
64
+ "step": 42
65
+ },
66
+ {
67
+ "epoch": 1.07,
68
+ "learning_rate": 0.0001354800547756731,
69
+ "loss": 3.1208,
70
+ "step": 45
71
+ },
72
+ {
73
+ "epoch": 1.19,
74
+ "learning_rate": 0.00012527997991807724,
75
+ "loss": 2.8559,
76
+ "step": 50
77
+ },
78
+ {
79
+ "epoch": 1.31,
80
+ "learning_rate": 0.00010724375598316453,
81
+ "loss": 2.9927,
82
+ "step": 55
83
+ },
84
+ {
85
+ "epoch": 1.43,
86
+ "learning_rate": 8.386493606940326e-05,
87
+ "loss": 2.9799,
88
+ "step": 60
89
+ },
90
+ {
91
+ "epoch": 1.55,
92
+ "learning_rate": 5.8375700540314535e-05,
93
+ "loss": 2.9474,
94
+ "step": 65
95
+ },
96
+ {
97
+ "epoch": 1.67,
98
+ "learning_rate": 3.4300000000000054e-05,
99
+ "loss": 2.8929,
100
+ "step": 70
101
+ },
102
+ {
103
+ "epoch": 1.79,
104
+ "learning_rate": 1.496636030269317e-05,
105
+ "loss": 2.821,
106
+ "step": 75
107
+ },
108
+ {
109
+ "epoch": 1.9,
110
+ "learning_rate": 3.04770552307078e-06,
111
+ "loss": 2.7814,
112
+ "step": 80
113
+ },
114
+ {
115
+ "epoch": 2.0,
116
+ "eval_loss": 3.0470073223114014,
117
+ "eval_runtime": 2.6251,
118
+ "eval_samples_per_second": 20.952,
119
+ "eval_steps_per_second": 2.667,
120
+ "step": 84
121
  }
122
  ],
123
+ "max_steps": 84,
124
+ "num_train_epochs": 2,
125
+ "total_flos": 86226370560000.0,
126
  "trial_name": null,
127
  "trial_params": null
128
  }
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:74489e50cba909b350e219cdb2c55931c833346ab293b4d186155c74b5daf6cf
3
  size 2863
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cd271d0572a7a3468694d79a611ddcf794549420f744198cdf5961c723dc8d78
3
  size 2863