zephyr-7b-dpo-qlora / config.json
RedaAlami's picture
End of training
04e7632 verified
{
"_name_or_path": "TII-Frontier-Team/falcon3-3b-instruct",
"activation": "swiglu",
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 10,
"eos_token_id": 127,
"head_dim": 256,
"hidden_act": "silu",
"hidden_size": 3072,
"initializer_range": 0.02,
"intermediate_size": 9216,
"max_position_embeddings": 4096,
"mlp_bias": false,
"model_type": "llama",
"num_attention_heads": 12,
"num_hidden_layers": 22,
"num_key_value_heads": 4,
"output_hidden_states": true,
"pad_token_id": 11,
"parallel_attn": false,
"pretraining_tp": 1,
"quantization_config": {
"_load_in_4bit": true,
"_load_in_8bit": false,
"bnb_4bit_compute_dtype": "bfloat16",
"bnb_4bit_quant_storage": "uint8",
"bnb_4bit_quant_type": "nf4",
"bnb_4bit_use_double_quant": false,
"llm_int8_enable_fp32_cpu_offload": false,
"llm_int8_has_fp16_weight": false,
"llm_int8_skip_modules": null,
"llm_int8_threshold": 6.0,
"load_in_4bit": true,
"load_in_8bit": false,
"quant_method": "bitsandbytes"
},
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 1000042,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.45.1",
"use_cache": true,
"vocab_size": 131072
}