Model: CaffeineThief/ttp_sft_kanana-1.5_steps_tram-step1-seed43 Source: Original Platform
31 lines
715 B
JSON
31 lines
715 B
JSON
{
|
|
"architectures": [
|
|
"LlamaForCausalLM"
|
|
],
|
|
"attention_bias": false,
|
|
"attention_dropout": 0.0,
|
|
"bos_token_id": 128000,
|
|
"eos_token_id": 128009,
|
|
"head_dim": 128,
|
|
"hidden_act": "silu",
|
|
"hidden_size": 1792,
|
|
"initializer_range": 0.02,
|
|
"intermediate_size": 8064,
|
|
"max_position_embeddings": 32768,
|
|
"mlp_bias": false,
|
|
"model_type": "llama",
|
|
"num_attention_heads": 24,
|
|
"num_hidden_layers": 32,
|
|
"num_key_value_heads": 8,
|
|
"pad_token_id": 128001,
|
|
"pretraining_tp": 1,
|
|
"rms_norm_eps": 1e-05,
|
|
"rope_scaling": null,
|
|
"rope_theta": 8000000.0,
|
|
"tie_word_embeddings": true,
|
|
"torch_dtype": "float32",
|
|
"transformers_version": "4.55.2",
|
|
"use_cache": false,
|
|
"vocab_size": 128259
|
|
}
|