Nethermind/Mpt-Instruct-DotNet-XS
046
1{2 "alibi": true,3 "alibi_bias_max": 8,4 "architectures": [5 "MosaicGPT"6 ],7 "attn_clip_qkv": null,8 "attn_impl": "torch",9 "attn_pdrop": 0,10 "attn_qk_ln": true,11 "attn_uses_sequence_id": false,12 "auto_map": {13 "AutoConfig": "configuration_mosaic_gpt.MosaicGPTConfig",14 "AutoModelForCausalLM": "mosaic_gpt.MosaicGPT"15 },16 "d_model": 2048,17 "emb_init_std": null,18 "emb_init_uniform_lim": null,19 "emb_pdrop": 0,20 "embedding_fraction": 1.0,21 "fan_mode": "fan_in",22 "init_device": "cpu",23 "init_div_is_residual": true,24 "init_gain": 0,25 "init_nonlinearity": "relu",26 "init_std": 0.02,27 "logit_scale": null,28 "low_precision_layernorm": true,29 "max_seq_len": 2048,30 "mlp_ratio": 4,31 "model_type": "mosaic_gpt",32 "n_heads": 16,33 "n_layers": 24,34 "no_bias": true,35 "param_init_fn": "kaiming_normal_",36 "prefix_lm": false,37 "resid_pdrop": 0,38 "softmax_scale": null,39 "tokenizer_name": "EleutherAI/gpt-neox-20b",40 "torch_dtype": "bfloat16",41 "transformers_version": "4.33.0",42 "use_cache": false,43 "verbose": 0,44 "vocab_size": 5043245}46 