TroyDoesAI's picture
1 Epoch of RAG Needle in Haystack on pruned vanilla model, removed 2.2 billion uneeded parameters.
782701c verified
raw
history blame
673 Bytes
{
"_name_or_path": ".\\Codestral-20B-RAG-Pruned\\",
"architectures": [
"MistralForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 6144,
"initializer_range": 0.02,
"intermediate_size": 16384,
"max_position_embeddings": 32768,
"model_type": "mistral",
"num_attention_heads": 48,
"num_hidden_layers": 48,
"num_key_value_heads": 8,
"rms_norm_eps": 1e-05,
"rope_theta": 1000000.0,
"sliding_window": null,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.40.2",
"use_cache": true,
"vocab_size": 32768
}