• unsloth를 이용해 ipc pdf를 전처리/파싱한 raw data로 pretraining한 모델

  • max_seq_length = 2000 -lora_rank=64, lora_alpha=64, lora_dropout=0

  • target_modules = ["q_proj", "k_proj", "v_proj", "o_proj",

  •     "gate_proj", "up_proj", "down_proj",
    
  •         "lm_head", "embed_tokens",]
    
Downloads last month
9
Safetensors
Model size
8.03B params
Tensor type
F32
·
FP16
·
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for sde0119/ipc-llama3-8b-ins-lora-pretrain-merged

Finetuned
(1425)
this model
Finetunes
3 models