{ | |
"_name_or_path": "openbuddy-qwen2.5llamaify-14b-v23.1-200k", | |
"architectures": [ | |
"LlamaForCausalLM" | |
], | |
"attention_bias": false, | |
"attention_dropout": 0.0, | |
"bos_token_id": 151643, | |
"eos_token_id": 151668, | |
"head_dim": 128, | |
"hidden_act": "silu", | |
"hidden_size": 5120, | |
"initializer_range": 0.02, | |
"intermediate_size": 13824, | |
"max_position_embeddings": 200000, | |
"mlp_bias": false, | |
"model_type": "llama", | |
"num_attention_heads": 40, | |
"num_hidden_layers": 48, | |
"num_key_value_heads": 8, | |
"pretraining_tp": 1, | |
"rms_norm_eps": 1e-05, | |
"rope_scaling": { | |
"factor": 8.0, | |
"high_freq_factor": 4.0, | |
"low_freq_factor": 1.0, | |
"original_max_position_embeddings": 32768, | |
"rope_type": "llama3" | |
}, | |
"rope_theta": 1000000.0, | |
"tie_word_embeddings": false, | |
"torch_dtype": "bfloat16", | |
"transformers_version": "4.45.0.dev0", | |
"use_cache": true, | |
"vocab_size": 152064 | |
} | |