{
    "vocab_size": 151665,
    "max_position_embeddings": 131072,
    "hidden_size": 3584,
    "intermediate_size": 18944,
    "num_hidden_layers": 28,
    "num_attention_heads": 28,
    "use_sliding_window": false,
    "sliding_window": null,
    "max_window_layers": 28,
    "num_key_value_heads": 4,
    "hidden_act": "silu",
    "initializer_range": 0.02,
    "rms_norm_eps": 1e-06,
    "use_cache": true,
    "rope_theta": 1000000.0,
    "rope_scaling": null,
    "attention_dropout": 0.0,
    "return_dict": true,
    "output_hidden_states": false,
    "output_attentions": false,
    "torchscript": false,
    "torch_dtype": "bfloat16",
    "use_bfloat16": false,
    "tf_legacy_loss": false,
    "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,
    "max_length": 20,
    "min_length": 0,
    "do_sample": false,
    "early_stopping": false,
    "num_beams": 1,
    "num_beam_groups": 1,
    "diversity_penalty": 0.0,
    "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,
    "architectures": [
        "Qwen2ForCausalLM"
    ],
    "finetuning_task": null,
    "id2label": {
        "0": "LABEL_0",
        "1": "LABEL_1"
    },
    "label2id": {
        "LABEL_0": 0,
        "LABEL_1": 1
    },
    "tokenizer_class": null,
    "prefix": null,
    "bos_token_id": 151643,
    "pad_token_id": 151643,
    "eos_token_id": 151643,
    "sep_token_id": null,
    "decoder_start_token_id": null,
    "task_specific_params": null,
    "problem_type": null,
    "_name_or_path": "/app/models/qwen",
    "_attn_implementation_autoset": true,
    "transformers_version": "4.48.3",
    "model_type": "qwen2",
    "use_mrope": false,
    "experiment_config": "log_path=PosixPath('train_output') seed=42 trainer_settings=TrainerSettings(eval_strategy='steps', save_strategy='no', per_device_train_batch_size=1, per_device_eval_batch_size=1, gradient_accumulation_steps=16, eval_steps=0.25, save_steps=150, logging_steps=1, learning_rate=3e-06, num_train_epochs=1, max_steps=-1, lr_scheduler_type='linear', lr_scheduler_kwargs={}, warmup_steps=0, warmup_ratio=0.03, fp16=False, bf16=True, tf32=False, torch_compile=False, optim='adamw_torch', adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-12, weight_decay=0.0, max_grad_norm=2.0, deepspeed=PosixPath('configs/exp/deepspeed/ds_config_stage_2.json'), save_total_limit=1, save_only_model=True, no_cuda=False, prediction_loss_only=False, load_best_model_at_end=False, logging_first_step=True, fsdp_config=None, fsdp='', dataloader_num_workers=12, dataloader_prefetch_factor=None, dataloader_persistent_workers=False, dataloader_pin_memory=True, gradient_checkpointing=False, gradient_checkpointing_kwargs={'use_reentrant': False}, neftune_noise_alpha=None, report_to=[]) tokenizer_settings=TokenizerSettings(use_fast=True, trust_remote_code=False, tokenizer_path=None, tokenizer_kwargs={}) special_tokens_settings=SpecialTokensSettings(bos_token='<|endoftext|>', eos_token='<|im_end|>', pad_token='<|endoftext|>', unk_token=None, sep_token=None) model_settings=PreTrainedModelSettings(model_path=PosixPath('/app/models/qwen'), model_type=<ModelType.CAUSAL: 'causal'>, model_kwargs={'attn_implementation': 'flash_attention_2'}, transformers_settings=ModelTransformersSettings(load_in_8bit=False, low_cpu_mem_usage=False, omit_base_model_save=None, trust_remote_code=False, torch_dtype='bfloat16'), resize_token_embeddings=True, embeddings_initialization_strategy=None, liger_kernels_settings=LigerKernelSettings(use_rope=True, use_cross_entropy=False, use_fused_linear_cross_entropy=True, use_mlp=True, use_rms_norm=True)) train_dataset_settings=ChatMultiDatasetSettings(sources=[DatasetSourceSettings(name='uch_train', system_prompt='You are a helpful assistant.', sample_rate=1.0, num_samples=None, records_path=PosixPath('data/uch/train_chat.jsonl'), records_data=None, offset=None, n_rows=None)], dataset_type=<DatasetType.CHAT: 'chat'>, only_last_replica_loss=True, only_answer_loss=True, random_cut=False, random_cut_seed=42, keep_end=True, max_tokens_count=4096, prompt_template=ChatPromptTemplate(prefix_template='<|im_start|>{role}\\n', suffix_template='<|im_end|>\\n', role_tag_mapping={'bot': 'assistant', 'user': 'user', 'system': 'system'}), ignore_system_prompt=False) val_dataset_settings=ChatMultiDatasetSettings(sources=[DatasetSourceSettings(name='uch_val', system_prompt='You are a helpful assistant.', sample_rate=1.0, num_samples=None, records_path=PosixPath('data/uch/val_chat500.jsonl'), records_data=None, offset=None, n_rows=None)], dataset_type=<DatasetType.CHAT: 'chat'>, only_last_replica_loss=True, only_answer_loss=True, random_cut=False, random_cut_seed=42, keep_end=True, max_tokens_count=4096, prompt_template=ChatPromptTemplate(prefix_template='<|im_start|>{role}\\n', suffix_template='<|im_end|>\\n', role_tag_mapping={'bot': 'assistant', 'user': 'user', 'system': 'system'}), ignore_system_prompt=False) logging_settings=WandbSettings(project_name='rm_hacking', run_name='qwen25-7b/uch/sft/lr-3e_06/1-ep', entity='turbo-alignment', tags=['qwen25-7b', 'sft'], notes=None, mode=<WandbMode.ONLINE: 'online'>) checkpoint_uploader_callback_parameters=None cherry_pick_settings=None"
}