Hugging Face's logo

Paladiso
/
66034ee5-a54d-46e2-b0b2-b10a4d504ab5

PEFT
Safetensors
llama
axolotl
Generated from Trainer
Model card Files
xet
 Community
Paladiso commited on
Commit
747af51
·
verified ·
1 Parent(s): 3a0abd7

End of training

Browse files
Files changed (1) hide show
  1. README.md +16 -19
README.md CHANGED
@@ -1,13 +1,13 @@
1
  ---
2
  library_name: peft
3
- base_model: samoline/c374ea20-5026-48ff-abab-3892b0b88cff
4
  tags:
5
  - axolotl
6
  - generated_from_trainer
7
  datasets:
8
- - Paladiso/dataset_551340a2-0886-4fde-9e57-97a233c286c6
9
  model-index:
10
- - name: 5abca899-5168-4061-94bf-3fd27d61cb75
11
  results: []
12
  ---
13
 
@@ -20,18 +20,17 @@ should probably proofread and complete it, then remove this comment. -->
20
  axolotl version: `0.6.0`
21
  ```yaml
22
  adapter: lora
23
- base_model: samoline/c374ea20-5026-48ff-abab-3892b0b88cff
24
  bf16: auto
25
  chat_template: llama3
26
  dataset_prepared_path: /workspace/axolotl/data/prepared
27
  datasets:
28
  - ds_type: json
29
  format: custom
30
- path: Paladiso/dataset_551340a2-0886-4fde-9e57-97a233c286c6
31
  type:
32
- field_input: input
33
- field_instruction: instruction
34
- field_output: output
35
  system_format: '{system}'
36
  system_prompt: ''
37
  debug: null
@@ -47,7 +46,7 @@ fsdp_config: null
47
  gradient_accumulation_steps: 4
48
  gradient_checkpointing: false
49
  group_by_length: false
50
- hub_model_id: Paladiso/5abca899-5168-4061-94bf-3fd27d61cb75
51
  hub_private_repo: true
52
  hub_repo: null
53
  hub_strategy: checkpoint
@@ -78,8 +77,6 @@ sample_packing: false
78
  save_safetensors: true
79
  saves_per_epoch: 4
80
  sequence_len: 512
81
- special_tokens:
82
- pad_token: <|endoftext|>
83
  strict: false
84
  tf32: false
85
  tokenizer_type: AutoTokenizer
@@ -89,10 +86,10 @@ use_accelerate: true
89
  val_set_size: 0.05
90
  wandb_entity: null
91
  wandb_mode: online
92
- wandb_name: 551340a2-0886-4fde-9e57-97a233c286c6
93
  wandb_project: Gradients-On-Demand
94
  wandb_run: your_name
95
- wandb_runid: 551340a2-0886-4fde-9e57-97a233c286c6
96
  warmup_steps: 10
97
  weight_decay: 0.0
98
  xformers_attention: null
@@ -101,11 +98,11 @@ xformers_attention: null
101
 
102
  </details><br>
103
 
104
- # 5abca899-5168-4061-94bf-3fd27d61cb75
105
 
106
- This model is a fine-tuned version of [samoline/c374ea20-5026-48ff-abab-3892b0b88cff](https://huggingface.co/samoline/c374ea20-5026-48ff-abab-3892b0b88cff) on the Paladiso/dataset_551340a2-0886-4fde-9e57-97a233c286c6 dataset.
107
  It achieves the following results on the evaluation set:
108
- - Loss: 1.1332
109
 
110
  ## Model description
111
 
@@ -139,9 +136,9 @@ The following hyperparameters were used during training:
139
 
140
  | Training Loss | Epoch | Step | Validation Loss |
141
  |:-------------:|:------:|:----:|:---------------:|
142
- | 1.1408 | 0.0006 | 3 | 1.1924 |
143
- | 1.0953 | 0.0012 | 6 | 1.1541 |
144
- | 0.9541 | 0.0018 | 9 | 1.1332 |
145
 
146
 
147
  ### Framework versions
 
1
  ---
2
  library_name: peft
3
+ base_model: lmsys/vicuna-7b-v1.3
4
  tags:
5
  - axolotl
6
  - generated_from_trainer
7
  datasets:
8
+ - Paladiso/dataset_c45a78c6-ae82-4ab6-9199-3b8fb9781583
9
  model-index:
10
+ - name: 66034ee5-a54d-46e2-b0b2-b10a4d504ab5
11
  results: []
12
  ---
13
 
 
20
  axolotl version: `0.6.0`
21
  ```yaml
22
  adapter: lora
23
+ base_model: lmsys/vicuna-7b-v1.3
24
  bf16: auto
25
  chat_template: llama3
26
  dataset_prepared_path: /workspace/axolotl/data/prepared
27
  datasets:
28
  - ds_type: json
29
  format: custom
30
+ path: Paladiso/dataset_c45a78c6-ae82-4ab6-9199-3b8fb9781583
31
  type:
32
+ field_instruction: inputs
33
+ field_output: targets
 
34
  system_format: '{system}'
35
  system_prompt: ''
36
  debug: null
 
46
  gradient_accumulation_steps: 4
47
  gradient_checkpointing: false
48
  group_by_length: false
49
+ hub_model_id: Paladiso/66034ee5-a54d-46e2-b0b2-b10a4d504ab5
50
  hub_private_repo: true
51
  hub_repo: null
52
  hub_strategy: checkpoint
 
77
  save_safetensors: true
78
  saves_per_epoch: 4
79
  sequence_len: 512
 
 
80
  strict: false
81
  tf32: false
82
  tokenizer_type: AutoTokenizer
 
86
  val_set_size: 0.05
87
  wandb_entity: null
88
  wandb_mode: online
89
+ wandb_name: c45a78c6-ae82-4ab6-9199-3b8fb9781583
90
  wandb_project: Gradients-On-Demand
91
  wandb_run: your_name
92
+ wandb_runid: c45a78c6-ae82-4ab6-9199-3b8fb9781583
93
  warmup_steps: 10
94
  weight_decay: 0.0
95
  xformers_attention: null
 
98
 
99
  </details><br>
100
 
101
+ # 66034ee5-a54d-46e2-b0b2-b10a4d504ab5
102
 
103
+ This model is a fine-tuned version of [lmsys/vicuna-7b-v1.3](https://huggingface.co/lmsys/vicuna-7b-v1.3) on the Paladiso/dataset_c45a78c6-ae82-4ab6-9199-3b8fb9781583 dataset.
104
  It achieves the following results on the evaluation set:
105
+ - Loss: 1.1334
106
 
107
  ## Model description
108
 
 
136
 
137
  | Training Loss | Epoch | Step | Validation Loss |
138
  |:-------------:|:------:|:----:|:---------------:|
139
+ | 1.1922 | 0.0316 | 3 | 1.2776 |
140
+ | 1.1629 | 0.0632 | 6 | 1.2478 |
141
+ | 1.1993 | 0.0947 | 9 | 1.1334 |
142
 
143
 
144
  ### Framework versions