doberst commited on
Commit
1ce5528
1 Parent(s): eef4a9d

Upload config.json

Browse files
Files changed (1) hide show
  1. config.json +89 -0
config.json ADDED
@@ -0,0 +1,89 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "aib_version": "model_archive_122723_tiny_llama-1.1b-29",
3
+ "training_dataset": [
4
+ "aib_label_samples_524s.jsonl"
5
+ ],
6
+ "training_timestamp": "Wed Dec 27 17:43:21 2023",
7
+ "training_comments": "tiny-llama-1.1b-27",
8
+ "vocab_size": 32000,
9
+ "max_position_embeddings": 2048,
10
+ "hidden_size": 2048,
11
+ "intermediate_size": 5632,
12
+ "num_hidden_layers": 22,
13
+ "num_attention_heads": 32,
14
+ "num_key_value_heads": 4,
15
+ "hidden_act": "silu",
16
+ "initializer_range": 0.02,
17
+ "rms_norm_eps": 1e-05,
18
+ "pretraining_tp": 1,
19
+ "use_cache": true,
20
+ "rope_theta": 10000.0,
21
+ "rope_scaling": null,
22
+ "attention_bias": false,
23
+ "attention_dropout": 0.0,
24
+ "return_dict": true,
25
+ "output_hidden_states": false,
26
+ "output_attentions": false,
27
+ "torchscript": false,
28
+ "torch_dtype": "float32",
29
+ "use_bfloat16": false,
30
+ "tf_legacy_loss": false,
31
+ "pruned_heads": {},
32
+ "tie_word_embeddings": false,
33
+ "is_encoder_decoder": false,
34
+ "is_decoder": false,
35
+ "cross_attention_hidden_size": null,
36
+ "add_cross_attention": false,
37
+ "tie_encoder_decoder": false,
38
+ "max_length": 20,
39
+ "min_length": 0,
40
+ "do_sample": false,
41
+ "early_stopping": false,
42
+ "num_beams": 1,
43
+ "num_beam_groups": 1,
44
+ "diversity_penalty": 0.0,
45
+ "temperature": 1.0,
46
+ "top_k": 50,
47
+ "top_p": 1.0,
48
+ "typical_p": 1.0,
49
+ "repetition_penalty": 1.0,
50
+ "length_penalty": 1.0,
51
+ "no_repeat_ngram_size": 0,
52
+ "encoder_no_repeat_ngram_size": 0,
53
+ "bad_words_ids": null,
54
+ "num_return_sequences": 1,
55
+ "chunk_size_feed_forward": 0,
56
+ "output_scores": false,
57
+ "return_dict_in_generate": false,
58
+ "forced_bos_token_id": null,
59
+ "forced_eos_token_id": null,
60
+ "remove_invalid_values": false,
61
+ "exponential_decay_length_penalty": null,
62
+ "suppress_tokens": null,
63
+ "begin_suppress_tokens": null,
64
+ "architectures": [
65
+ "LlamaForCausalLM"
66
+ ],
67
+ "finetuning_task": null,
68
+ "id2label": {
69
+ "0": "LABEL_0",
70
+ "1": "LABEL_1"
71
+ },
72
+ "label2id": {
73
+ "LABEL_0": 0,
74
+ "LABEL_1": 1
75
+ },
76
+ "tokenizer_class": null,
77
+ "prefix": null,
78
+ "bos_token_id": 1,
79
+ "pad_token_id": null,
80
+ "eos_token_id": 2,
81
+ "sep_token_id": null,
82
+ "decoder_start_token_id": null,
83
+ "task_specific_params": null,
84
+ "problem_type": null,
85
+ "_name_or_path": "TinyLlama/TinyLlama-1.1B-intermediate-step-1195k-token-2.5T",
86
+ "transformers_version": "4.36.1",
87
+ "model_type": "llama",
88
+ "trained": "custom training"
89
+ }