| { |
| "return_dict": true, |
| "output_hidden_states": false, |
| "torchscript": false, |
| "dtype": null, |
| "pruned_heads": {}, |
| "tie_word_embeddings": false, |
| "chunk_size_feed_forward": 0, |
| "is_encoder_decoder": false, |
| "is_decoder": false, |
| "cross_attention_hidden_size": null, |
| "add_cross_attention": false, |
| "tie_encoder_decoder": false, |
| "architectures": ["RecursiveMaskedLM"], |
| "finetuning_task": null, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1" |
| }, |
| "label2id": { |
| "LABEL_0": 0, |
| "LABEL_1": 1 |
| }, |
| "task_specific_params": null, |
| "problem_type": null, |
| "tokenizer_class": null, |
| "prefix": null, |
| "bos_token_id": null, |
| "pad_token_id": null, |
| "eos_token_id": null, |
| "sep_token_id": null, |
| "decoder_start_token_id": null, |
| "max_length": 20, |
| "min_length": 0, |
| "do_sample": false, |
| "early_stopping": false, |
| "num_beams": 1, |
| "temperature": 1.0, |
| "top_k": 50, |
| "top_p": 1.0, |
| "typical_p": 1.0, |
| "repetition_penalty": 1.0, |
| "length_penalty": 1.0, |
| "no_repeat_ngram_size": 0, |
| "encoder_no_repeat_ngram_size": 0, |
| "bad_words_ids": null, |
| "num_return_sequences": 1, |
| "output_scores": false, |
| "return_dict_in_generate": false, |
| "forced_bos_token_id": null, |
| "forced_eos_token_id": null, |
| "remove_invalid_values": false, |
| "exponential_decay_length_penalty": null, |
| "suppress_tokens": null, |
| "begin_suppress_tokens": null, |
| "num_beam_groups": 1, |
| "diversity_penalty": 0.0, |
| "_name_or_path": "", |
| "transformers_version": "4.57.0", |
| "tf_legacy_loss": false, |
| "use_bfloat16": false, |
| "base_model_config": { |
| "return_dict": true, |
| "output_hidden_states": false, |
| "torchscript": false, |
| "dtype": "bfloat16", |
| "pruned_heads": {}, |
| "tie_word_embeddings": false, |
| "chunk_size_feed_forward": 0, |
| "is_encoder_decoder": false, |
| "is_decoder": false, |
| "cross_attention_hidden_size": null, |
| "add_cross_attention": false, |
| "tie_encoder_decoder": false, |
| "architectures": [ |
| "LLaDAModelLM" |
| ], |
| "finetuning_task": null, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1" |
| }, |
| "label2id": { |
| "LABEL_0": 0, |
| "LABEL_1": 1 |
| }, |
| "task_specific_params": null, |
| "problem_type": null, |
| "tokenizer_class": null, |
| "prefix": null, |
| "bos_token_id": null, |
| "pad_token_id": 76, |
| "eos_token_id": 76, |
| "sep_token_id": null, |
| "decoder_start_token_id": null, |
| "max_length": 20, |
| "min_length": 0, |
| "do_sample": false, |
| "early_stopping": false, |
| "num_beams": 1, |
| "temperature": 1.0, |
| "top_k": 50, |
| "top_p": 1.0, |
| "typical_p": 1.0, |
| "repetition_penalty": 1.0, |
| "length_penalty": 1.0, |
| "no_repeat_ngram_size": 0, |
| "encoder_no_repeat_ngram_size": 0, |
| "bad_words_ids": null, |
| "num_return_sequences": 1, |
| "output_scores": false, |
| "return_dict_in_generate": false, |
| "forced_bos_token_id": null, |
| "forced_eos_token_id": null, |
| "remove_invalid_values": false, |
| "exponential_decay_length_penalty": null, |
| "suppress_tokens": null, |
| "begin_suppress_tokens": null, |
| "num_beam_groups": 1, |
| "diversity_penalty": 0.0, |
| "_name_or_path": "Fraser/LLaDA-8B-Base-gg2m", |
| "transformers_version": "4.57.0", |
| "d_model": 4096, |
| "n_heads": 32, |
| "n_kv_heads": 32, |
| "n_layers": 32, |
| "mlp_ratio": 4, |
| "mlp_hidden_size": 12288, |
| "activation_type": "silu", |
| "block_type": "llama", |
| "block_group_size": 1, |
| "alibi": false, |
| "alibi_bias_max": 8.0, |
| "rope": true, |
| "rope_full_precision": true, |
| "flash_attention": false, |
| "attention_dropout": 0.0, |
| "multi_query_attention": null, |
| "attention_layer_norm": false, |
| "residual_dropout": 0.0, |
| "embedding_dropout": 0.0, |
| "input_emb_norm": false, |
| "layer_norm_type": "rms", |
| "layer_norm_with_affine": true, |
| "rms_norm_eps": 1e-05, |
| "attention_layer_norm_with_affine": true, |
| "max_sequence_length": 4096, |
| "rope_theta": 500000.0, |
| "include_qkv_bias": false, |
| "include_bias": false, |
| "bias_for_layer_norm": false, |
| "scale_logits": false, |
| "vocab_size": 85, |
| "embedding_size": 85, |
| "weight_tying": false, |
| "mask_token_id": 78, |
| "init_device": "meta", |
| "init_fn": "mitchell", |
| "init_std": 0.02, |
| "init_cutoff_factor": null, |
| "precision": "amp_bf16", |
| "auto_map": { |
| "AutoConfig": "configuration_llada.LLaDAConfig", |
| "AutoModelForCausalLM": "modeling_llada.LLaDAModelLM", |
| "AutoModel": "modeling_llada.LLaDAModelLM" |
| }, |
| "model_type": "llada", |
| "use_cache": false, |
| "tf_legacy_loss": false, |
| "use_bfloat16": false, |
| "output_attentions": false |
| }, |
| "num_recursions": 4, |
| "normalization": "softmax", |
| "loss_weight": "linear", |
| "mask_token_id": 78, |
| "gradient_steps": null, |
| "schedule": "linear", |
| "causal_strength": 1.0, |
| "temperature_max": 0.0, |
| "entropy_target_max": 0.0, |
| "entropy_floor_max": 0.0, |
| "smear_sigma_max": 0.0, |
| "noise_std_max": 0.0, |
| "iteration_rope_dim_fraction": 0.0, |
| "use_recursion_checkpointing": true, |
| "soft_embedding_method": "softmax", |
| "soft_embedding_ema_step": 1.0, |
| "flow_matching_enabled": false, |
| "flow_matching_lambda": 0.5, |
| "flow_matching_t_distribution": "logit_normal", |
| "flow_matching_t_logit_mean": -0.4, |
| "flow_matching_t_logit_std": 1.0, |
| "flow_matching_t_min": 0.01, |
| "flow_matching_t_max": 0.99, |
| "flow_matching_noise_scale": 2.0, |
| "flow_matching_mask_scale": false, |
| "self_distillation_enabled": false, |
| "self_distillation_lambda": 0.5, |
| "self_distillation_temperature_min": 1.5, |
| "self_distillation_temperature_max": 10.0, |
| "self_distillation_temperature_distribution": "log_uniform", |
| "self_distillation_teacher": "first", |
| "model_type": "recursive-mlm", |
| "output_attentions": false, |
| "auto_map": { |
| "AutoConfig": "configuration_recursive.RecursiveMLMConfig", |
| "AutoModel": "modeling_recursive.RecursiveMaskedLM" |
| } |
| } |
