Upload model trained with transformers trainer on nqzfaizal77ai/base-random-imitation-en-1bc-dataset,cw 384,lr 1e-4,3 epoch,batch size 4,cosine with restart
dc358a5
verified
| { | |
| "architectures": [ | |
| "MistralForCausalLM" | |
| ], | |
| "attention_dropout": 0.1, | |
| "bos_token_id": 1, | |
| "eos_token_id": 2, | |
| "head_dim": 96, | |
| "hidden_act": "silu", | |
| "hidden_size": 1024, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 4096, | |
| "max_position_embeddings": 8192, | |
| "model_type": "mistral", | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 10, | |
| "num_key_value_heads": 4, | |
| "rms_norm_eps": 1e-05, | |
| "rope_theta": 1000000.0, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "float16", | |
| "transformers_version": "4.51.3", | |
| "use_cache": true, | |
| "vocab_size": 32768 | |
| } | |