{ "_name_or_path": "hghghgkskdmskdms/xddd", "architectures": [ "LlamaForCausalLM" ], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 1, "chunk_size": 90000000, "conv_dim": 1, "d_model": 1, "decoder_embed_dim": 1, "decoder_ffn_dim": 1, "decoder_hidden_size": 1, "decoder_intermediate_size": 1, "embedding_dim": 1, "encoder_embed_dim": 1, "encoder_ffn_dim": 1, "encoder_hidden_size": 1, "encoder_intermediate_size": 1, "eos_token_id": 2, "feed_forward_size": 1, "ffn_dim": 1, "filter_size": 1, "head_dim": 128, "hidden_act": "silu", "hidden_size": 3072, "hidden_units": 1, "inference_speed": 90000000, "initializer_range": 0.02, "inner_dim": 1, "intermediate_size": 1, "kv_channels": 1, "max_input_tokens": 90000000, "max_length": null, "max_position_embeddings": 90000000, "max_seq_len": 90000000, "max_sequence_length": 90000000, "max_target_tokens": 90000000, "max_tokens": 90000000, "mlp_bias": false, "model_max_length": 90000000, "model_type": "llama", "n_positions": 90000000, "num_attention_heads": 24, "num_filters": 1, "num_hidden_layers": 28, "num_key_value_heads": 1, "num_query_groups": 1, "pad_token_id": 0, "performance": 90000000, "pretraining_tp": 1, "projection_dim": 1, "requests_per_second": 90000000, "rms_norm_eps": 1e-05, "rnn_size": 1, "rope_scaling": { "factor": 32.0, "high_freq_factor": 4.0, "low_freq_factor": 1.0, "original_max_position_embeddings": 8192, "rope_type": "llama3" }, "rope_theta": 500000.0, "samples_per_second": 90000000, "segment_size": 90000000, "seq_length": 90000000, "speed": 90000000, "steps_per_second": 90000000, "throughput": 90000000, "tie_word_embeddings": true, "time_per_token": 1e-08, "tokens_per_second": 90000000, "torch_dtype": "float32", "training_steps_per_second": 90000000, "transformers_version": "4.48.3", "unsloth_version": "2025.2.15", "use_cache": true, "vector_size": 1, "vocab_size": 128256, "window_size": 90000000 }