Views
No views yet
0.4.11base_model: google/gemma-2b-it
2model_type: AutoModelForCausalLM
3tokenizer_type: AutoTokenizer
4# tokenizer_config: google/gemma-2b-it
5
6load_in_8bit: false
7load_in_4bit: false
8strict: false
9
10datasets:
11 - path: /scratch/jiarui14/dpo-ood/LLM-finetune/WX-Reward-Modeling/pair-pm/ultrafeedback-single
12 conversation: gemma
13 type: sharegpt.load_ultrachat
14 split: "train"
15 train_on_split: "train"
16
17warmup_steps: 40
18val_set_size: 0.0
19output_dir: ./pm_models/gemma-2b-it_lr1e-5_ultrafeedback
20#wandb_project: preference-models
21#wandb_entity: domain-generalization
22wandb_watch:
23wandb_name: "gemma-2b-it_lr1e-5_ultrafeedback"
24#_response_only
25wandb_log_model:
26
27train_on_inputs: false
28
29save_safetensors: true
30#noisy_embedding_alpha: 10.0 # default for sharegpt type
31dataset_prepared_path: data/gemma-2b-it/ultrafeedback
32
33
34dataset_processes: 48
35#torch_compile: true
36sequence_len: 3072
37sample_packing: true
38pad_to_sequence_len: true
39
40trust_remote_code: True
41adapter:
42lora_model_dir:
43
44
45
46
47gradient_checkpointing: false
48
49#warmup_ratio: 0.1
50gradient_accumulation_steps: 8
51micro_batch_size: 2
52num_epochs: 1
53optimizer: adamw_torch_fused
54lr_scheduler: cosine
55learning_rate: 1e-5
56
57weight_decay: 0.0
58max_grad_norm: 1.0
59
60
61group_by_length: false
62bf16: auto
63fp16: false
64tf32: true
65
66early_stopping_patience:
67local_rank:
68logging_steps: 1
69xformers_attention:
70flash_attention: true
71
72
73eval_steps:
74eval_table_size:
75eval_table_max_new_tokens:
76save_steps: 50
77save_strategy: "steps"
78save_total_limit: 2
79debug:
80
81
82ddp: #true
83deepspeed: #deepspeed/zero1.json # multi-gpu only
84
85fsdp:
86fsdp_config:
87special_tokens:
88
89