dsakerkwq commited on
Commit
1ea479d
1 Parent(s): 7fbbd5b

End of training

Browse files
Files changed (2) hide show
  1. README.md +153 -0
  2. adapter_model.bin +3 -0
README.md ADDED
@@ -0,0 +1,153 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: peft
3
+ license: apache-2.0
4
+ base_model: unsloth/Qwen2-0.5B
5
+ tags:
6
+ - axolotl
7
+ - generated_from_trainer
8
+ model-index:
9
+ - name: 4616dd30-0309-4496-adc2-62fe53420323
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: unsloth/Qwen2-0.5B
23
+ bf16: true
24
+ chat_template: llama3
25
+ dataset_prepared_path: null
26
+ datasets:
27
+ - data_files:
28
+ - 1e7bb670723eee7d_train_data.json
29
+ ds_type: json
30
+ format: custom
31
+ path: /workspace/input_data/1e7bb670723eee7d_train_data.json
32
+ type:
33
+ field_input: source
34
+ field_instruction: prompt
35
+ field_output: response
36
+ format: '{instruction} {input}'
37
+ no_input_format: '{instruction}'
38
+ system_format: '{system}'
39
+ system_prompt: ''
40
+ debug: null
41
+ deepspeed: null
42
+ eval_max_new_tokens: 128
43
+ eval_table_size: null
44
+ evals_per_epoch: 4
45
+ flash_attention: false
46
+ fp16: false
47
+ fsdp: null
48
+ fsdp_config: null
49
+ gradient_accumulation_steps: 8
50
+ gradient_checkpointing: true
51
+ group_by_length: true
52
+ hub_model_id: dsakerkwq/4616dd30-0309-4496-adc2-62fe53420323
53
+ learning_rate: 0.0002
54
+ load_in_4bit: false
55
+ load_in_8bit: false
56
+ lora_alpha: 32
57
+ lora_dropout: 0.05
58
+ lora_fan_in_fan_out: null
59
+ lora_model_dir: null
60
+ lora_r: 16
61
+ lora_target_linear: true
62
+ lr_scheduler: cosine
63
+ max_memory:
64
+ 0: 75GiB
65
+ max_steps: 30
66
+ micro_batch_size: 2
67
+ mlflow_experiment_name: /tmp/1e7bb670723eee7d_train_data.json
68
+ model_type: AutoModelForCausalLM
69
+ num_epochs: 3
70
+ optimizer: adamw_torch
71
+ output_dir: miner_id_24
72
+ pad_to_sequence_len: true
73
+ s2_attention: false
74
+ sample_packing: false
75
+ saves_per_epoch: 4
76
+ sequence_len: 2048
77
+ strict: false
78
+ tf32: true
79
+ tokenizer_type: AutoTokenizer
80
+ train_on_inputs: false
81
+ trust_remote_code: true
82
+ val_set_size: 0.05
83
+ wandb_entity: null
84
+ wandb_mode: online
85
+ wandb_name: 4616dd30-0309-4496-adc2-62fe53420323
86
+ wandb_project: Gradients-On-Demand
87
+ wandb_runid: 4616dd30-0309-4496-adc2-62fe53420323
88
+ warmup_steps: 100
89
+ weight_decay: 0.01
90
+ xformers_attention: false
91
+
92
+ ```
93
+
94
+ </details><br>
95
+
96
+ # 4616dd30-0309-4496-adc2-62fe53420323
97
+
98
+ This model is a fine-tuned version of [unsloth/Qwen2-0.5B](https://huggingface.co/unsloth/Qwen2-0.5B) on the None dataset.
99
+ It achieves the following results on the evaluation set:
100
+ - Loss: nan
101
+
102
+ ## Model description
103
+
104
+ More information needed
105
+
106
+ ## Intended uses & limitations
107
+
108
+ More information needed
109
+
110
+ ## Training and evaluation data
111
+
112
+ More information needed
113
+
114
+ ## Training procedure
115
+
116
+ ### Training hyperparameters
117
+
118
+ The following hyperparameters were used during training:
119
+ - learning_rate: 0.0002
120
+ - train_batch_size: 2
121
+ - eval_batch_size: 2
122
+ - seed: 42
123
+ - gradient_accumulation_steps: 8
124
+ - total_train_batch_size: 16
125
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
126
+ - lr_scheduler_type: cosine
127
+ - lr_scheduler_warmup_steps: 100
128
+ - training_steps: 30
129
+
130
+ ### Training results
131
+
132
+ | Training Loss | Epoch | Step | Validation Loss |
133
+ |:-------------:|:------:|:----:|:---------------:|
134
+ | 0.0 | 0.0020 | 1 | nan |
135
+ | 0.0 | 0.0059 | 3 | nan |
136
+ | 0.0 | 0.0118 | 6 | nan |
137
+ | 0.0 | 0.0177 | 9 | nan |
138
+ | 0.0 | 0.0236 | 12 | nan |
139
+ | 0.0 | 0.0295 | 15 | nan |
140
+ | 0.0 | 0.0354 | 18 | nan |
141
+ | 0.0 | 0.0413 | 21 | nan |
142
+ | 0.0 | 0.0472 | 24 | nan |
143
+ | 0.0 | 0.0531 | 27 | nan |
144
+ | 0.0 | 0.0590 | 30 | nan |
145
+
146
+
147
+ ### Framework versions
148
+
149
+ - PEFT 0.13.2
150
+ - Transformers 4.46.0
151
+ - Pytorch 2.5.0+cu124
152
+ - Datasets 3.0.1
153
+ - Tokenizers 0.20.1
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a74249ec9396e9432efbf96fc2579c9498769989264104fcbfdffe8ac29a8f76
3
+ size 35313738