jettjaniak commited on
Commit
311aef1
1 Parent(s): 82c5eaa

Upload folder using huggingface_hub

Browse files
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 0,
8
+ "eos_token_id": 1,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 12,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 48,
13
+ "max_position_embeddings": 512,
14
+ "model_type": "llama",
15
+ "num_attention_heads": 2,
16
+ "num_hidden_layers": 1,
17
+ "num_key_value_heads": 1,
18
+ "pad_token_id": 2,
19
+ "pretraining_tp": 1,
20
+ "rms_norm_eps": 1e-06,
21
+ "rope_scaling": null,
22
+ "rope_theta": 10000.0,
23
+ "tie_word_embeddings": false,
24
+ "torch_dtype": "float32",
25
+ "transformers_version": "4.40.0",
26
+ "use_cache": true,
27
+ "vocab_size": 4096
28
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 1,
5
+ "pad_token_id": 2,
6
+ "transformers_version": "4.40.0"
7
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ea6d2f409eb01bcb03bb4f21ff3cc7d12463810c2ded3ba70c12a15be209c965
3
+ size 403216
optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1f31bf8dba98637205d3ead8f53e54987aaba212244df5819ef4e5c07b6a5e0
3
+ size 813674
run_context.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "device": "cuda",
3
+ "gpu_name": "Quadro M4000",
4
+ "torch_version": "2.1.2+cu121",
5
+ "delphi_version": "0.2",
6
+ "transformers_version": "4.40.0",
7
+ "os": "#164-Ubuntu SMP Tue Mar 21 14:23:17 UTC 2023"
8
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<bos>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<eos>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<pad>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ }
23
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<bos>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<eos>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "<pad>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ }
27
+ },
28
+ "bos_token": "<bos>",
29
+ "clean_up_tokenization_spaces": true,
30
+ "eos_token": "<eos>",
31
+ "model_max_length": 1000000000000000019884624838656,
32
+ "pad_token": "<pad>",
33
+ "tokenizer_class": "PreTrainedTokenizerFast"
34
+ }
training_config.json ADDED
@@ -0,0 +1,72 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "model_config": {
3
+ "model_class": "LlamaForCausalLM",
4
+ "vocab_size": 4096,
5
+ "hidden_act": "silu",
6
+ "max_position_embeddings": 512,
7
+ "initializer_range": 0.02,
8
+ "rms_norm_eps": 1e-06,
9
+ "bos_token_id": 0,
10
+ "eos_token_id": 1,
11
+ "pad_token_id": 2,
12
+ "tie_word_embeddings": false,
13
+ "rope_theta": 10000.0,
14
+ "rope_scaling": null,
15
+ "attention_bias": false,
16
+ "attention_dropout": 0.0,
17
+ "hidden_size": 12,
18
+ "intermediate_size": 48,
19
+ "num_attention_heads": 2,
20
+ "num_hidden_layers": 1,
21
+ "num_key_value_heads": 1
22
+ },
23
+ "max_seq_len": 512,
24
+ "run_name": "2024_05_24_13_29_02",
25
+ "out_dir": "/home/paperspace/.local/share/delphi/2024_05_24_13_29_02",
26
+ "device": "auto",
27
+ "checkpoint_interval": 400,
28
+ "extra_checkpoint_iters": [
29
+ 1,
30
+ 2,
31
+ 4,
32
+ 8,
33
+ 16,
34
+ 32,
35
+ 64,
36
+ 128,
37
+ 256,
38
+ 512
39
+ ],
40
+ "log_interval": 40,
41
+ "eval_iters": 10,
42
+ "resume_from_path": null,
43
+ "batch_size": 256,
44
+ "max_epochs": 1,
45
+ "grad_clip": 1.0,
46
+ "gradient_accumulation_steps": 4,
47
+ "adam": {
48
+ "learning_rate": 0.0005,
49
+ "weight_decay": 0.1,
50
+ "beta1": 0.9,
51
+ "beta2": 0.95,
52
+ "decay_lr": true,
53
+ "warmup_iters": 1000,
54
+ "min_lr": 0.0
55
+ },
56
+ "batch_ordering_seed": 1337,
57
+ "torch_seed": 42,
58
+ "save_optimizer": true,
59
+ "dataset": {
60
+ "name": "delphi-suite/stories-tokenized",
61
+ "feature": "tokens",
62
+ "train_split": "train[:3%]",
63
+ "validation_split": "validation"
64
+ },
65
+ "tokenizer": "delphi-suite/stories-tokenizer",
66
+ "wandb": "jettjaniak/delphi",
67
+ "out_repo": "jettjaniak/stories-llama2-100k-v02-test2",
68
+ "debug_config": {
69
+ "no_training": false,
70
+ "no_eval": false
71
+ }
72
+ }
training_state.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "iter_num": 125,
3
+ "lr": 6.2e-05,
4
+ "epoch": 0,
5
+ "step": 124
6
+ }