llama.cpp/nllb_testing/results/model_config.json

16 lines
394 B
JSON

{
"model_name": "facebook/nllb-200-distilled-600M",
"d_model": 1024,
"encoder_layers": 12,
"decoder_layers": 12,
"encoder_attention_heads": 16,
"decoder_attention_heads": 16,
"encoder_ffn_dim": 4096,
"decoder_ffn_dim": 4096,
"max_position_embeddings": 1024,
"vocab_size": 256204,
"bos_token_id": 0,
"eos_token_id": 2,
"pad_token_id": 1,
"decoder_start_token_id": 2
}