adarshxs commited on
Commit
c39defb
1 Parent(s): 71582fe
README.md ADDED
@@ -0,0 +1,143 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - generated_from_trainer
4
+ model-index:
5
+ - name: out
6
+ results: []
7
+ ---
8
+
9
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
10
+ should probably proofread and complete it, then remove this comment. -->
11
+
12
+ [<img src="https://raw.githubusercontent.com/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
13
+ <details><summary>See axolotl config</summary>
14
+
15
+ axolotl version: `0.3.0`
16
+ ```yaml
17
+ base_model: ./TinyLlama-1.1B-intermediate-step-1431k-3T
18
+
19
+ model_type: LlamaForCausalLM
20
+ tokenizer_type: LlamaTokenizer
21
+ is_llama_derived_model: true
22
+
23
+ load_in_8bit: false
24
+ load_in_4bit: false
25
+ strict: false
26
+
27
+ datasets:
28
+ - path: ./openhermes
29
+ type: alpaca
30
+ dataset_prepared_path:
31
+ val_set_size: 0.05
32
+ output_dir: ./out
33
+
34
+ sequence_len: 4096
35
+ sample_packing: false
36
+
37
+ adapter:
38
+ lora_model_dir:
39
+ lora_r:
40
+ lora_alpha:
41
+ lora_dropout:
42
+ lora_target_linear:
43
+ lora_fan_in_fan_out:
44
+
45
+ wandb_project: tinyllama-openhermes
46
+ wandb_entity: tensoic
47
+ wandb_watch:
48
+ wandb_name:
49
+ wandb_log_model:
50
+
51
+ gradient_accumulation_steps: 2
52
+ micro_batch_size: 8
53
+ num_epochs: 1
54
+ optimizer: adamw_bnb_8bit
55
+ lr_scheduler: cosine
56
+ learning_rate: 0.0002
57
+
58
+ train_on_inputs: false
59
+ group_by_length: false
60
+ bf16: false
61
+ fp16: true
62
+ tf32: false
63
+
64
+ gradient_checkpointing: true
65
+ early_stopping_patience:
66
+ resume_from_checkpoint:
67
+ local_rank:
68
+ logging_steps: 1
69
+ xformers_attention: true
70
+ flash_attention:
71
+
72
+ warmup_steps: 100
73
+ evals_per_epoch: 4
74
+ eval_table_size:
75
+ saves_per_epoch: 1
76
+ debug:
77
+ deepspeed: zero2.json
78
+ weight_decay: 0.0
79
+ fsdp:
80
+ fsdp_config:
81
+ special_tokens:
82
+ bos_token: "<s>"
83
+ eos_token: "</s>"
84
+ unk_token: "<unk>"
85
+
86
+ ```
87
+
88
+ </details><br>
89
+
90
+ # out
91
+
92
+ This model was trained from scratch on the None dataset.
93
+ It achieves the following results on the evaluation set:
94
+ - Loss: 7.4061
95
+
96
+ ## Model description
97
+
98
+ More information needed
99
+
100
+ ## Intended uses & limitations
101
+
102
+ More information needed
103
+
104
+ ## Training and evaluation data
105
+
106
+ More information needed
107
+
108
+ ## Training procedure
109
+
110
+ ### Training hyperparameters
111
+
112
+ The following hyperparameters were used during training:
113
+ - learning_rate: 0.0002
114
+ - train_batch_size: 8
115
+ - eval_batch_size: 8
116
+ - seed: 42
117
+ - distributed_type: multi-GPU
118
+ - num_devices: 8
119
+ - gradient_accumulation_steps: 2
120
+ - total_train_batch_size: 128
121
+ - total_eval_batch_size: 64
122
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
123
+ - lr_scheduler_type: cosine
124
+ - lr_scheduler_warmup_steps: 100
125
+ - num_epochs: 1
126
+ - mixed_precision_training: Native AMP
127
+
128
+ ### Training results
129
+
130
+ | Training Loss | Epoch | Step | Validation Loss |
131
+ |:-------------:|:-----:|:----:|:---------------:|
132
+ | 3.0006 | 0.0 | 1 | 1.6838 |
133
+ | 0.855 | 0.25 | 451 | 1.5228 |
134
+ | 6.8636 | 0.5 | 902 | 7.4147 |
135
+ | 6.9346 | 0.75 | 1353 | 7.4061 |
136
+
137
+
138
+ ### Framework versions
139
+
140
+ - Transformers 4.36.2
141
+ - Pytorch 2.0.1+cu117
142
+ - Datasets 2.15.0
143
+ - Tokenizers 0.15.0
config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "./TinyLlama-1.1B-intermediate-step-1431k-3T",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 5632,
14
+ "max_position_embeddings": 4096,
15
+ "model_type": "llama",
16
+ "num_attention_heads": 32,
17
+ "num_hidden_layers": 22,
18
+ "num_key_value_heads": 4,
19
+ "pretraining_tp": 1,
20
+ "rms_norm_eps": 1e-05,
21
+ "rope_scaling": null,
22
+ "rope_theta": 10000.0,
23
+ "tie_word_embeddings": false,
24
+ "torch_dtype": "float16",
25
+ "transformers_version": "4.36.2",
26
+ "use_cache": false,
27
+ "vocab_size": 32000
28
+ }
generation_config.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token_id": 1,
3
+ "eos_token_id": 2,
4
+ "max_length": 2048,
5
+ "pad_token_id": 0,
6
+ "transformers_version": "4.36.2"
7
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94c785c532bd545ba4e239d6c0f096cc13750c8130b8f5a8592b40048a6ede59
3
+ size 2200123773
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
3
+ size 499723
tokenizer_config.json ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "bos_token": "<s>",
31
+ "clean_up_tokenization_spaces": false,
32
+ "eos_token": "</s>",
33
+ "legacy": false,
34
+ "model_max_length": 1000000000000000019884624838656,
35
+ "pad_token": "</s>",
36
+ "padding_side": "right",
37
+ "sp_model_kwargs": {},
38
+ "spaces_between_special_tokens": false,
39
+ "tokenizer_class": "LlamaTokenizer",
40
+ "trust_remote_code": false,
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false,
43
+ "use_fast": true
44
+ }