jeiku commited on
Commit
a56df09
1 Parent(s): 84737ee

Upload 10 files

Browse files
README.md ADDED
@@ -0,0 +1,133 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - generated_from_trainer
6
+ base_model: Qwen/Qwen2-7B
7
+ model-index:
8
+ - name: outputs/out
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ [<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)
16
+ <details><summary>See axolotl config</summary>
17
+
18
+ axolotl version: `0.4.1`
19
+ ```yaml
20
+ adapter: qlora
21
+ base_model: Qwen/Qwen2-7B
22
+ bf16: auto
23
+ dataset_prepared_path: null
24
+ datasets:
25
+ - path: ResplendentAI/Synthetic_Soul_1k
26
+ type: alpaca
27
+ debug: null
28
+ deepspeed: null
29
+ early_stopping_patience: null
30
+ eval_sample_packing: false
31
+ evals_per_epoch: 2
32
+ flash_attention: true
33
+ fp16: null
34
+ gradient_accumulation_steps: 4
35
+ gradient_checkpointing: true
36
+ gradient_checkpointing_kwargs:
37
+ use_reentrant: false
38
+ group_by_length: false
39
+ learning_rate: 2.0e-05
40
+ load_in_4bit: true
41
+ load_in_8bit: false
42
+ local_rank: null
43
+ logging_steps: 1
44
+ lora_alpha: 128
45
+ lora_dropout: 0.05
46
+ lora_fan_in_fan_out: null
47
+ lora_model_dir: null
48
+ lora_r: 128
49
+ lora_target_linear: true
50
+ lr_scheduler: cosine
51
+ micro_batch_size: 4
52
+ num_epochs: 4
53
+ optimizer: adamw_torch
54
+ output_dir: ./outputs/out
55
+ pad_to_sequence_len: false
56
+ resume_from_checkpoint: null
57
+ sample_packing: false
58
+ saves_per_epoch: 1
59
+ sequence_len: 8192
60
+ special_tokens: null
61
+ strict: false
62
+ tf32: true
63
+ train_on_inputs: false
64
+ trust_remote_code: true
65
+ val_set_size: 0.05
66
+ wandb_entity: null
67
+ wandb_log_model: null
68
+ wandb_name: null
69
+ wandb_project: null
70
+ wandb_watch: null
71
+ warmup_steps: 10
72
+ weight_decay: 0.0
73
+ xformers_attention: null
74
+
75
+ ```
76
+
77
+ </details><br>
78
+
79
+ # outputs/out
80
+
81
+ This model is a fine-tuned version of [Qwen/Qwen2-7B](https://huggingface.co/Qwen/Qwen2-7B) on the None dataset.
82
+ It achieves the following results on the evaluation set:
83
+ - Loss: 1.4422
84
+
85
+ ## Model description
86
+
87
+ More information needed
88
+
89
+ ## Intended uses & limitations
90
+
91
+ More information needed
92
+
93
+ ## Training and evaluation data
94
+
95
+ More information needed
96
+
97
+ ## Training procedure
98
+
99
+ ### Training hyperparameters
100
+
101
+ The following hyperparameters were used during training:
102
+ - learning_rate: 2e-05
103
+ - train_batch_size: 4
104
+ - eval_batch_size: 4
105
+ - seed: 42
106
+ - gradient_accumulation_steps: 4
107
+ - total_train_batch_size: 16
108
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
109
+ - lr_scheduler_type: cosine
110
+ - lr_scheduler_warmup_steps: 10
111
+ - num_epochs: 4
112
+
113
+ ### Training results
114
+
115
+ | Training Loss | Epoch | Step | Validation Loss |
116
+ |:-------------:|:------:|:----:|:---------------:|
117
+ | 1.8248 | 0.0168 | 1 | 1.7354 |
118
+ | 1.2323 | 0.5042 | 30 | 1.4900 |
119
+ | 1.2644 | 1.0084 | 60 | 1.4405 |
120
+ | 1.1181 | 1.5126 | 90 | 1.4438 |
121
+ | 1.0902 | 2.0168 | 120 | 1.4244 |
122
+ | 1.0422 | 2.5210 | 150 | 1.4514 |
123
+ | 1.0635 | 3.0252 | 180 | 1.4363 |
124
+ | 0.9551 | 3.5294 | 210 | 1.4422 |
125
+
126
+
127
+ ### Framework versions
128
+
129
+ - PEFT 0.11.1
130
+ - Transformers 4.41.1
131
+ - Pytorch 2.1.2+cu121
132
+ - Datasets 2.19.1
133
+ - Tokenizers 0.19.1
adapter_config.json ADDED
@@ -0,0 +1,34 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "Qwen/Qwen2-7B",
5
+ "bias": "none",
6
+ "fan_in_fan_out": null,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layer_replication": null,
10
+ "layers_pattern": null,
11
+ "layers_to_transform": null,
12
+ "loftq_config": {},
13
+ "lora_alpha": 128,
14
+ "lora_dropout": 0.05,
15
+ "megatron_config": null,
16
+ "megatron_core": "megatron.core",
17
+ "modules_to_save": null,
18
+ "peft_type": "LORA",
19
+ "r": 128,
20
+ "rank_pattern": {},
21
+ "revision": null,
22
+ "target_modules": [
23
+ "up_proj",
24
+ "down_proj",
25
+ "q_proj",
26
+ "gate_proj",
27
+ "k_proj",
28
+ "o_proj",
29
+ "v_proj"
30
+ ],
31
+ "task_type": "CAUSAL_LM",
32
+ "use_dora": false,
33
+ "use_rslora": false
34
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e39bb8e9f5da59c5413b36ce65ca392002b13d3c4675c74676793d386e12867
3
+ size 646064554
added_tokens.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "<|endoftext|>": 151643,
3
+ "<|im_end|>": 151645,
4
+ "<|im_start|>": 151644
5
+ }
config.json ADDED
@@ -0,0 +1,42 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "Qwen/Qwen2-7B",
3
+ "architectures": [
4
+ "Qwen2ForCausalLM"
5
+ ],
6
+ "attention_dropout": 0.0,
7
+ "eos_token_id": 151643,
8
+ "hidden_act": "silu",
9
+ "hidden_size": 3584,
10
+ "initializer_range": 0.02,
11
+ "intermediate_size": 18944,
12
+ "max_position_embeddings": 131072,
13
+ "max_window_layers": 28,
14
+ "model_type": "qwen2",
15
+ "num_attention_heads": 28,
16
+ "num_hidden_layers": 28,
17
+ "num_key_value_heads": 4,
18
+ "quantization_config": {
19
+ "_load_in_4bit": true,
20
+ "_load_in_8bit": false,
21
+ "bnb_4bit_compute_dtype": "bfloat16",
22
+ "bnb_4bit_quant_storage": "bfloat16",
23
+ "bnb_4bit_quant_type": "nf4",
24
+ "bnb_4bit_use_double_quant": true,
25
+ "llm_int8_enable_fp32_cpu_offload": false,
26
+ "llm_int8_has_fp16_weight": false,
27
+ "llm_int8_skip_modules": null,
28
+ "llm_int8_threshold": 6.0,
29
+ "load_in_4bit": true,
30
+ "load_in_8bit": false,
31
+ "quant_method": "bitsandbytes"
32
+ },
33
+ "rms_norm_eps": 1e-06,
34
+ "rope_theta": 1000000.0,
35
+ "sliding_window": 131072,
36
+ "tie_word_embeddings": false,
37
+ "torch_dtype": "bfloat16",
38
+ "transformers_version": "4.41.1",
39
+ "use_cache": false,
40
+ "use_sliding_window": false,
41
+ "vocab_size": 152064
42
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
special_tokens_map.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|im_start|>",
4
+ "<|im_end|>"
5
+ ],
6
+ "eos_token": {
7
+ "content": "<|endoftext|>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false
12
+ },
13
+ "pad_token": {
14
+ "content": "<|endoftext|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false
19
+ }
20
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "151643": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": false,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ },
12
+ "151644": {
13
+ "content": "<|im_start|>",
14
+ "lstrip": false,
15
+ "normalized": false,
16
+ "rstrip": false,
17
+ "single_word": false,
18
+ "special": true
19
+ },
20
+ "151645": {
21
+ "content": "<|im_end|>",
22
+ "lstrip": false,
23
+ "normalized": false,
24
+ "rstrip": false,
25
+ "single_word": false,
26
+ "special": true
27
+ }
28
+ },
29
+ "additional_special_tokens": [
30
+ "<|im_start|>",
31
+ "<|im_end|>"
32
+ ],
33
+ "bos_token": null,
34
+ "chat_template": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
35
+ "clean_up_tokenization_spaces": false,
36
+ "eos_token": "<|endoftext|>",
37
+ "errors": "replace",
38
+ "model_max_length": 32768,
39
+ "pad_token": "<|endoftext|>",
40
+ "split_special_tokens": false,
41
+ "tokenizer_class": "Qwen2Tokenizer",
42
+ "unk_token": null
43
+ }
vocab.json ADDED
The diff for this file is too large to render. See raw diff