File size: 1,402 Bytes
5f27f9a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 |
{
"_name_or_path": "google/recurrentgemma-9b",
"architectures": [
"RecurrentGemmaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"attention_window_size": 2048,
"block_types": [
"recurrent",
"recurrent",
"attention"
],
"bos_token_id": 2,
"conv1d_width": 4,
"eos_token_id": 1,
"final_w_init_variance_scale": 0.05263157894736842,
"head_dim": 256,
"hidden_activation": "gelu_pytorch_tanh",
"hidden_size": 4096,
"intermediate_size": 24576,
"logits_soft_cap": 30.0,
"lru_width": 4096,
"model_type": "recurrent_gemma",
"num_attention_heads": 16,
"num_hidden_layers": 38,
"num_key_value_heads": 1,
"pad_token_id": 0,
"partial_rotary_factor": 0.5,
"quantization_config": {
"_load_in_4bit": true,
"_load_in_8bit": false,
"bnb_4bit_compute_dtype": "bfloat16",
"bnb_4bit_quant_storage": "uint8",
"bnb_4bit_quant_type": "nf4",
"bnb_4bit_use_double_quant": true,
"llm_int8_enable_fp32_cpu_offload": false,
"llm_int8_has_fp16_weight": false,
"llm_int8_skip_modules": null,
"llm_int8_threshold": 6.0,
"load_in_4bit": true,
"load_in_8bit": false,
"quant_method": "bitsandbytes"
},
"rms_norm_eps": 1e-06,
"rope_theta": 10000.0,
"torch_dtype": "bfloat16",
"transformers_version": "4.42.0.dev0",
"use_cache": true,
"vocab_size": 256000,
"w_init_variance_scale": 0.01
}
|