lesso13 commited on
Commit
3f22aec
·
verified ·
1 Parent(s): fd78e9a

End of training

Browse files
Files changed (2) hide show
  1. README.md +164 -0
  2. adapter_model.bin +3 -0
README.md ADDED
@@ -0,0 +1,164 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: llama3
4
+ base_model: DeepMount00/Llama-3-8b-Ita
5
+ tags:
6
+ - axolotl
7
+ - generated_from_trainer
8
+ model-index:
9
+ - name: cdd77e3c-d251-4a54-bfca-dbd43f9a1725
10
+ results: []
11
+ ---
12
+
13
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
14
+ should probably proofread and complete it, then remove this comment. -->
15
+
16
+ [<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
17
+ <details><summary>See axolotl config</summary>
18
+
19
+ axolotl version: `0.4.1`
20
+ ```yaml
21
+ adapter: lora
22
+ auto_find_batch_size: true
23
+ base_model: DeepMount00/Llama-3-8b-Ita
24
+ bf16: auto
25
+ chat_template: llama3
26
+ dataset_prepared_path: null
27
+ datasets:
28
+ - data_files:
29
+ - 8bb3de2515f68e34_train_data.json
30
+ ds_type: json
31
+ format: custom
32
+ path: /workspace/input_data/8bb3de2515f68e34_train_data.json
33
+ type:
34
+ field_instruction: instruction
35
+ field_output: output
36
+ format: '{instruction}'
37
+ no_input_format: '{instruction}'
38
+ system_format: '{system}'
39
+ system_prompt: ''
40
+ debug: null
41
+ deepspeed: null
42
+ do_eval: true
43
+ early_stopping_patience: 3
44
+ eval_max_new_tokens: 128
45
+ eval_steps: 50
46
+ evals_per_epoch: null
47
+ flash_attention: true
48
+ fp16: false
49
+ fsdp: null
50
+ fsdp_config: null
51
+ gradient_accumulation_steps: 2
52
+ gradient_checkpointing: false
53
+ group_by_length: true
54
+ hub_model_id: lesso13/cdd77e3c-d251-4a54-bfca-dbd43f9a1725
55
+ hub_repo: null
56
+ hub_strategy: checkpoint
57
+ hub_token: null
58
+ learning_rate: 0.000213
59
+ load_in_4bit: false
60
+ load_in_8bit: false
61
+ local_rank: null
62
+ logging_steps: 10
63
+ lora_alpha: 128
64
+ lora_dropout: 0.05
65
+ lora_fan_in_fan_out: null
66
+ lora_model_dir: null
67
+ lora_r: 64
68
+ lora_target_linear: true
69
+ lr_scheduler: cosine
70
+ max_grad_norm: 1.0
71
+ max_steps: 500
72
+ micro_batch_size: 4
73
+ mlflow_experiment_name: /tmp/G.O.D/8bb3de2515f68e34_train_data.json
74
+ model_type: AutoModelForCausalLM
75
+ num_epochs: 1
76
+ optimizer: adamw_bnb_8bit
77
+ output_dir: miner_id_24
78
+ pad_to_sequence_len: true
79
+ resume_from_checkpoint: null
80
+ s2_attention: null
81
+ sample_packing: false
82
+ save_steps: 50
83
+ saves_per_epoch: null
84
+ sequence_len: 512
85
+ special_tokens:
86
+ pad_token: <|eot_id|>
87
+ strict: false
88
+ tf32: true
89
+ tokenizer_type: AutoTokenizer
90
+ train_on_inputs: false
91
+ trust_remote_code: true
92
+ val_set_size: 0.05
93
+ wandb_entity: null
94
+ wandb_mode: online
95
+ wandb_name: 3402e0d8-6057-4b92-a376-02ad538dbab9
96
+ wandb_project: 13a
97
+ wandb_run: your_name
98
+ wandb_runid: 3402e0d8-6057-4b92-a376-02ad538dbab9
99
+ warmup_steps: 50
100
+ weight_decay: 0.0
101
+ xformers_attention: null
102
+
103
+ ```
104
+
105
+ </details><br>
106
+
107
+ # cdd77e3c-d251-4a54-bfca-dbd43f9a1725
108
+
109
+ This model is a fine-tuned version of [DeepMount00/Llama-3-8b-Ita](https://huggingface.co/DeepMount00/Llama-3-8b-Ita) on the None dataset.
110
+ It achieves the following results on the evaluation set:
111
+ - Loss: 0.1643
112
+
113
+ ## Model description
114
+
115
+ More information needed
116
+
117
+ ## Intended uses & limitations
118
+
119
+ More information needed
120
+
121
+ ## Training and evaluation data
122
+
123
+ More information needed
124
+
125
+ ## Training procedure
126
+
127
+ ### Training hyperparameters
128
+
129
+ The following hyperparameters were used during training:
130
+ - learning_rate: 0.000213
131
+ - train_batch_size: 4
132
+ - eval_batch_size: 4
133
+ - seed: 42
134
+ - gradient_accumulation_steps: 2
135
+ - total_train_batch_size: 8
136
+ - optimizer: Use OptimizerNames.ADAMW_BNB with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
137
+ - lr_scheduler_type: cosine
138
+ - lr_scheduler_warmup_steps: 50
139
+ - training_steps: 500
140
+
141
+ ### Training results
142
+
143
+ | Training Loss | Epoch | Step | Validation Loss |
144
+ |:-------------:|:------:|:----:|:---------------:|
145
+ | No log | 0.0004 | 1 | 0.6095 |
146
+ | 0.2669 | 0.0213 | 50 | 0.2906 |
147
+ | 0.2541 | 0.0425 | 100 | 0.2186 |
148
+ | 0.275 | 0.0638 | 150 | 0.2177 |
149
+ | 0.2931 | 0.0851 | 200 | 0.2002 |
150
+ | 0.254 | 0.1063 | 250 | 0.1978 |
151
+ | 0.2914 | 0.1276 | 300 | 0.1880 |
152
+ | 0.1944 | 0.1488 | 350 | 0.1762 |
153
+ | 0.234 | 0.1701 | 400 | 0.1677 |
154
+ | 0.2486 | 0.1914 | 450 | 0.1651 |
155
+ | 0.2541 | 0.2126 | 500 | 0.1643 |
156
+
157
+
158
+ ### Framework versions
159
+
160
+ - PEFT 0.13.2
161
+ - Transformers 4.46.0
162
+ - Pytorch 2.5.0+cu124
163
+ - Datasets 3.0.1
164
+ - Tokenizers 0.20.1
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:00b46a96e398be4cb7a9578be71f17c707b43f10163f6adbef71d47a34c30a77
3
+ size 671250634