sergioalves commited on
Commit
8b900b4
·
verified ·
1 Parent(s): 979b683

End of training

Browse files
Files changed (2) hide show
  1. README.md +163 -0
  2. adapter_model.bin +3 -0
README.md ADDED
@@ -0,0 +1,163 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: apache-2.0
4
+ base_model: tiiuae/falcon-rw-1b
5
+ tags:
6
+ - axolotl
7
+ - generated_from_trainer
8
+ model-index:
9
+ - name: 41664e15-2771-4b94-8ea5-337be5b9d7b1
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
+ base_model: tiiuae/falcon-rw-1b
23
+ bf16: auto
24
+ chat_template: llama3
25
+ dataset_prepared_path: null
26
+ datasets:
27
+ - data_files:
28
+ - 07a28d376e7dbf1b_train_data.json
29
+ ds_type: json
30
+ format: custom
31
+ path: /workspace/input_data/07a28d376e7dbf1b_train_data.json
32
+ type:
33
+ field_input: negative
34
+ field_instruction: feature_clean
35
+ field_output: positive
36
+ format: '{instruction} {input}'
37
+ no_input_format: '{instruction}'
38
+ system_format: '{system}'
39
+ system_prompt: ''
40
+ debug: null
41
+ deepspeed: null
42
+ device: cuda
43
+ early_stopping_patience: 1
44
+ eval_max_new_tokens: 128
45
+ eval_steps: 5
46
+ eval_table_size: null
47
+ evals_per_epoch: null
48
+ flash_attention: false
49
+ fp16: null
50
+ gradient_accumulation_steps: 4
51
+ gradient_checkpointing: true
52
+ group_by_length: true
53
+ hub_model_id: sergioalves/41664e15-2771-4b94-8ea5-337be5b9d7b1
54
+ hub_repo: null
55
+ hub_strategy: checkpoint
56
+ hub_token: null
57
+ learning_rate: 0.0002
58
+ load_in_4bit: false
59
+ load_in_8bit: false
60
+ local_rank: null
61
+ logging_steps: 3
62
+ lora_alpha: 32
63
+ lora_dropout: 0.05
64
+ lora_fan_in_fan_out: null
65
+ lora_model_dir: null
66
+ lora_r: 16
67
+ lora_target_linear: true
68
+ lr_scheduler: cosine
69
+ max_memory:
70
+ 0: 79GiB
71
+ max_steps: 30
72
+ micro_batch_size: 4
73
+ mlflow_experiment_name: /tmp/07a28d376e7dbf1b_train_data.json
74
+ model_type: AutoModelForCausalLM
75
+ num_epochs: 1
76
+ optim_args:
77
+ adam_beta1: 0.9
78
+ adam_beta2: 0.95
79
+ adam_epsilon: 1e-5
80
+ optimizer: adamw_torch
81
+ output_dir: miner_id_24
82
+ pad_to_sequence_len: true
83
+ resume_from_checkpoint: null
84
+ s2_attention: null
85
+ sample_packing: false
86
+ save_steps: 10
87
+ sequence_len: 1024
88
+ special_tokens:
89
+ pad_token: <|endoftext|>
90
+ strict: false
91
+ tf32: false
92
+ tokenizer_type: AutoTokenizer
93
+ train_on_inputs: true
94
+ trust_remote_code: true
95
+ val_set_size: 0.05
96
+ wandb_entity: null
97
+ wandb_mode: online
98
+ wandb_name: c2b3bfa1-01d0-4998-88b2-47cd6a64c629
99
+ wandb_project: Gradients-On-Demand
100
+ wandb_run: your_name
101
+ wandb_runid: c2b3bfa1-01d0-4998-88b2-47cd6a64c629
102
+ warmup_steps: 5
103
+ weight_decay: 0.001
104
+ xformers_attention: true
105
+
106
+ ```
107
+
108
+ </details><br>
109
+
110
+ # 41664e15-2771-4b94-8ea5-337be5b9d7b1
111
+
112
+ This model is a fine-tuned version of [tiiuae/falcon-rw-1b](https://huggingface.co/tiiuae/falcon-rw-1b) on the None dataset.
113
+ It achieves the following results on the evaluation set:
114
+ - Loss: 1.3190
115
+
116
+ ## Model description
117
+
118
+ More information needed
119
+
120
+ ## Intended uses & limitations
121
+
122
+ More information needed
123
+
124
+ ## Training and evaluation data
125
+
126
+ More information needed
127
+
128
+ ## Training procedure
129
+
130
+ ### Training hyperparameters
131
+
132
+ The following hyperparameters were used during training:
133
+ - learning_rate: 0.0002
134
+ - train_batch_size: 4
135
+ - eval_batch_size: 4
136
+ - seed: 42
137
+ - gradient_accumulation_steps: 4
138
+ - total_train_batch_size: 16
139
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=adam_beta1=0.9,adam_beta2=0.95,adam_epsilon=1e-5
140
+ - lr_scheduler_type: cosine
141
+ - lr_scheduler_warmup_steps: 5
142
+ - training_steps: 30
143
+
144
+ ### Training results
145
+
146
+ | Training Loss | Epoch | Step | Validation Loss |
147
+ |:-------------:|:------:|:----:|:---------------:|
148
+ | No log | 0.0053 | 1 | 1.6881 |
149
+ | 6.2728 | 0.0263 | 5 | 1.5987 |
150
+ | 6.0131 | 0.0526 | 10 | 1.4852 |
151
+ | 5.4887 | 0.0789 | 15 | 1.3929 |
152
+ | 5.4168 | 0.1053 | 20 | 1.3469 |
153
+ | 5.2769 | 0.1316 | 25 | 1.3215 |
154
+ | 5.1204 | 0.1579 | 30 | 1.3190 |
155
+
156
+
157
+ ### Framework versions
158
+
159
+ - PEFT 0.13.2
160
+ - Transformers 4.46.0
161
+ - Pytorch 2.5.0+cu124
162
+ - Datasets 3.0.1
163
+ - Tokenizers 0.20.1
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:756171b087c3cac46075a6f08ab3e725fb299f772da3060b2da54671ff3c589e
3
+ size 50402250