Text Generation
Transformers
English
codegen
Inference Endpoints
codegen-6B-lora / config.json
mhhmm's picture
Update config.json
ac49971
raw
history blame
675 Bytes
{
"base_model_name_or_path": "Salesforce/codegen-6B-mono",
"peft_model_id": "mhhmm/codegen-6B-LoRA",
"_name_or_path": "codegen-6B-mono",
"architectures": [
"CodeGenForCausalLM"
],
"tokenizer_class": "GPT2Tokenizer",
"attention_probs_dropout_prob": 0.0,
"attn_pdrop": 0.0,
"bos_token_id": 0,
"embd_pdrop": 0.0,
"eos_token_id": 0,
"model_type": "codegen",
"pad_token_id": 1023,
"fan_in_fan_out": false,
"inference_mode": true,
"init_lora_weights": true,
"lora_alpha": 32,
"lora_dropout": 0.05,
"modules_to_save": null,
"peft_type": "LORA",
"r": 16,
"target_modules": [
"q_proj",
"v_proj"
],
"task_type": "CAUSAL_LM"
}