Hugging Face's logo

Wilson-Wei2002
/
elo_type1

Safetensors
mistral
Model card Files
xet
Community
elo_type1
File size: 3,229 Bytes 
36bf776
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
 
architecture:
    backbone_dtype: bfloat16
    gradient_checkpointing: true
    intermediate_dropout: 0.0
    pretrained: true
    pretrained_weights: ''
augmentation:
    neftune_noise_alpha: 0.0
    random_parent_probability: 0.0
    skip_parent_probability: 0.0
    token_mask_probability: 0.0
dataset:
    add_eos_token_to_answer: true
    add_eos_token_to_prompt: true
    add_eos_token_to_system: true
    answer_column: chosen_response
    chatbot_author: H2O.ai
    chatbot_name: h2oGPT
    data_sample: 1.0
    data_sample_choice:
    - Train
    - Validation
    id_column: None
    limit_chained_samples: true
    mask_prompt_labels: true
    only_last_answer: false
    parent_id_column: None
    personalize: false
    prompt_column:
    - prompt
    prompt_column_separator: \n\n
    rejected_answer_column: rejected_response
    rejected_prompt_column: None
    system_column: None
    text_answer_separator: ''
    text_prompt_start: ''
    text_system_start: <|system|>
    train_dataframe: /root/h2o-llmstudio/data/user/
    validation_dataframe: None
    validation_size: 0.01
    validation_strategy: automatic
environment:
    compile_model: false
    deepspeed_allgather_bucket_size: 1000000
    deepspeed_method: ZeRO2
    deepspeed_reduce_bucket_size: 1000000
    deepspeed_stage3_param_persistence_threshold: 1000000
    deepspeed_stage3_prefetch_bucket_size: 1000000
    find_unused_parameters: false
    gpus:
    - '0'
    - '1'
    - '2'
    - '3'
    huggingface_branch: main
    mixed_precision: true
    mixed_precision_dtype: bfloat16
    number_of_workers: 8
    seed: 69
    trust_remote_code: true
    use_deepspeed: false
experiment_name: 
llm_backbone: 
logging:
    log_all_ranks: false
    log_step_size: absolute
    logger: Neptune
    neptune_project: nischaychai/nis-exp
    wandb_entity: ''
    wandb_project: ''
output_directory: /root/h2o-llmstudio/output/user/
prediction:
    batch_size_inference: 0
    do_sample: false
    max_length_inference: 256
    max_time: 0.0
    metric: GPT
    metric_gpt_model: gpt-3.5-turbo-0301
    metric_gpt_template: mt-bench
    min_length_inference: 2
    num_beams: 1
    num_history: 4
    repetition_penalty: 1.0
    stop_tokens: ''
    temperature: 0.0
    top_k: 0
    top_p: 1.0
problem_type: text_dpo_modeling
tokenizer:
    add_prompt_answer_tokens: false
    max_length: 1280
    padding_quantile: 1.0
    tokenizer_kwargs: '{"use_fast": true, "add_prefix_space": false}'
training:
    attention_implementation: auto
    batch_size: 2
    beta: 0.34
    differential_learning_rate: 1.0e-05
    differential_learning_rate_layers: []
    drop_last_batch: true
    epochs: 3
    evaluate_before_training: false
    evaluation_epochs: 1.0
    freeze_layers: []
    grad_accumulation: 4
    gradient_clip: 2.0
    learning_rate: 4.0e-06
    lora: true
    lora_alpha: 96
    lora_dropout: 0.0
    lora_r: 32
    lora_target_modules: ''
    lora_unfreeze_layers: []
    loss_function: SimPOLoss
    min_learning_rate_ratio: 0.0
    optimizer: AdamW
    save_checkpoint: last
    schedule: Cosine
    simpo_gamma: 1.0
    train_validation_data: false
    use_dora: false
    use_rslora: false
    warmup_epochs: 0.0
    weight_decay: 1.0e-05