PyJamba1B / config.json
isemmanuelolowe's picture
Training in progress, step 500
d8c354f verified
{
"_name_or_path": "PyJamba1B/checkpoint-265000",
"architectures": [
"JambaForCausalLM"
],
"attention_dropout": 0.0,
"attn_layer_offset": 4,
"attn_layer_period": 8,
"auto_map": {
"AutoConfig": "configuration_jamba.JambaConfig",
"AutoModel": "ai21labs/Jamba-v0.1--modeling_jamba.JambaModel",
"AutoModelForCausalLM": "modeling_jamba.JambaForCausalLM",
"AutoModelForSequenceClassification": "ai21labs/Jamba-v0.1--model.JambaForSequenceClassification"
},
"bos_token_id": 1,
"calc_logits_for_entire_prompt": false,
"eos_token_id": 2,
"expert_layer_offset": 1,
"expert_layer_period": 2,
"hidden_act": "silu",
"hidden_size": 1024,
"initializer_range": 0.02,
"intermediate_size": 2048,
"mamba_conv_bias": true,
"mamba_d_conv": 4,
"mamba_d_state": 16,
"mamba_dt_rank": 256,
"mamba_expand": 2,
"mamba_inner_layernorms": true,
"mamba_proj_bias": false,
"model_type": "jamba",
"n_ctx": 262144,
"num_attention_heads": 32,
"num_experts": 32,
"num_experts_per_tok": 2,
"num_hidden_layers": 8,
"num_key_value_heads": 8,
"output_router_logits": false,
"pad_token_id": 0,
"rms_norm_eps": 1e-06,
"router_aux_loss_coef": 0.001,
"sliding_window": null,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.40.0.dev0",
"use_cache": true,
"use_mamba_kernels": true,
"vocab_size": 65536
}