Text Generation
Transformers
Safetensors
mixtral
reasoning
preference_learning
nca
text-generation-inference
Inference Endpoints
Eurux-8x22b-kto / config.json
hanbin's picture
Upload folder using huggingface_hub (#1)
9e4788f verified
raw
history blame contribute delete
749 Bytes
{
"architectures": [
"MixtralForCausalLM"
],
"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": 65536,
"model_type": "mixtral",
"num_attention_heads": 48,
"num_experts_per_tok": 2,
"num_hidden_layers": 56,
"num_key_value_heads": 8,
"num_local_experts": 8,
"output_router_logits": false,
"rms_norm_eps": 1e-05,
"rope_theta": 1000000,
"router_aux_loss_coef": 0.001,
"router_jitter_noise": 0.0,
"sliding_window": null,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.40.0.dev0",
"use_cache": true,
"vocab_size": 32000
}