{ | |
"_name_or_path": "/gpfswork/rech/fof/commun/locost/locost-ckpt/pretrained", | |
"agg_strategy": "add", | |
"architectures": [ | |
"LOCOSTForConditionalGeneration" | |
], | |
"bidirectional": true, | |
"d_ff": 2048, | |
"d_kv": 64, | |
"d_model": 768, | |
"d_state": 512, | |
"decoder_start_token_id": 0, | |
"dense_act_fn": "gelu_new", | |
"dropout_rate": 0.0, | |
"encoder_attention_type": "local", | |
"eos_token_id": 1, | |
"feed_forward_proj": "gated-gelu", | |
"gating": true, | |
"global_block_size": 16, | |
"initializer_factor": 1.0, | |
"is_encoder_decoder": true, | |
"is_gated_act": true, | |
"layer_norm_epsilon": 1e-06, | |
"local_radius": 127, | |
"num_decoder_layers": 12, | |
"num_heads": 12, | |
"num_layers": 12, | |
"num_ssm_heads": 1, | |
"pad_token_id": 0, | |
"relative_attention_max_distance": 128, | |
"relative_attention_num_buckets": 32, | |
"torch_dtype": "float32", | |
"transformers_version": "4.27.1", | |
"use_cache": true, | |
"use_fast_fft_conv": true, | |
"vocab_size": 32100 | |
} | |