llama-global-mask-5724 / config.json
ricardomonti08's picture
Upload folder using huggingface_hub
f0644c2 verified
{
"architectures": [
"MPTForCausalLM"
],
"attn_config": {
"alibi": false,
"alibi_bias_max": 8,
"attn_impl": "torch",
"attn_logit_softcapping": null,
"attn_pdrop": 0.0,
"attn_type": "grouped_query_attention",
"attn_uses_sequence_id": true,
"clip_qkv": null,
"fused_qkv": true,
"kv_dim": null,
"kv_n_heads": 8,
"qk_gn": false,
"qk_ln": false,
"rope": true,
"rope_dail_config": {
"pos_idx_in_fp32": true,
"type": "original",
"xpos_scale_base": 512
},
"rope_hf_config": {
"factor": 1.0,
"type": "no_scaling"
},
"rope_impl": "dail",
"rope_theta": 50000,
"sliding_window_size": -1,
"softmax_scale": null
},
"auto_map": {
"AutoConfig": "configuration_mpt.MPTConfig",
"AutoModelForCausalLM": "modeling_mpt.MPTForCausalLM"
},
"block_overrides": null,
"d_model": 3072,
"emb_pdrop": 0.0,
"embedding_fraction": 1.0,
"expansion_ratio": 4,
"fc_type": {
"name": "torch"
},
"ffn_config": {
"fc_type": {
"name": "torch"
},
"ffn_act_fn": {
"name": "silu"
},
"ffn_type": "mptmlp"
},
"ffn_hidden_size": 8192,
"final_logit_softcapping": null,
"init_config": {
"emb_init_std": null,
"emb_init_uniform_lim": null,
"fan_mode": "fan_in",
"init_div_is_residual": true,
"init_gain": 0.0,
"init_nonlinearity": "relu",
"init_std": null,
"name": "kaiming_normal_"
},
"init_device": "cpu",
"layer_norm_epsilon": 1e-05,
"learned_pos_emb": false,
"logit_scale": null,
"max_seq_len": 4096,
"model_type": "mpt",
"n_heads": 24,
"n_layers": 28,
"no_bias": false,
"norm_eps": 1e-05,
"norm_type": "rmsnorm",
"resid_pdrop": 0.0,
"torch_dtype": "bfloat16",
"transformers_version": "4.51.3",
"use_cache": false,
"use_pad_tok_in_ffn": true,
"vocab_size": 152000
}