Add a6_coupled_proposer_coder round 4 Solver weights

Browse files

Files changed (15) hide show

.gitattributes +1 -0
a6_coupled_proposer_coder/round_4/args.json +556 -0
a6_coupled_proposer_coder/round_4/chat_template.jinja +154 -0
a6_coupled_proposer_coder/round_4/config.json +117 -0
a6_coupled_proposer_coder/round_4/generation_config.json +9 -0
a6_coupled_proposer_coder/round_4/model-00001-of-00004.safetensors +3 -0
a6_coupled_proposer_coder/round_4/model-00002-of-00004.safetensors +3 -0
a6_coupled_proposer_coder/round_4/model-00003-of-00004.safetensors +3 -0
a6_coupled_proposer_coder/round_4/model-00004-of-00004.safetensors +3 -0
a6_coupled_proposer_coder/round_4/model.safetensors.index.json +768 -0
a6_coupled_proposer_coder/round_4/preprocessor_config.json +21 -0
a6_coupled_proposer_coder/round_4/processor_config.json +61 -0
a6_coupled_proposer_coder/round_4/tokenizer.json +3 -0
a6_coupled_proposer_coder/round_4/tokenizer_config.json +33 -0
a6_coupled_proposer_coder/round_4/trainer_state.json +2454 -0

.gitattributes CHANGED Viewed

@@ -71,3 +71,4 @@ a1_api_proposer/round_3/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 a6_coupled_proposer_coder/round_3/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 a1_api_proposer/round_4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 c2_uniform_difficulty/round_3/tokenizer.json filter=lfs diff=lfs merge=lfs -text

 a6_coupled_proposer_coder/round_3/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 a1_api_proposer/round_4/tokenizer.json filter=lfs diff=lfs merge=lfs -text
 c2_uniform_difficulty/round_3/tokenizer.json filter=lfs diff=lfs merge=lfs -text
+a6_coupled_proposer_coder/round_4/tokenizer.json filter=lfs diff=lfs merge=lfs -text

a6_coupled_proposer_coder/round_4/args.json ADDED Viewed

	@@ -0,0 +1,556 @@

+{
+  "output_dir": "/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_004/solver",
+  "per_device_train_batch_size": 1,
+  "num_train_epochs": 3.0,
+  "max_steps": 160,
+  "learning_rate": 1e-06,
+  "lr_scheduler_type": "cosine",
+  "lr_scheduler_kwargs": null,
+  "warmup_steps": 0,
+  "optim": "adamw_torch_fused",
+  "optim_args": null,
+  "weight_decay": 0.1,
+  "adam_beta1": 0.9,
+  "adam_beta2": 0.95,
+  "adam_epsilon": 1e-08,
+  "optim_target_modules": null,
+  "gradient_accumulation_steps": 8,
+  "average_tokens_across_devices": true,
+  "max_grad_norm": 1.0,
+  "label_smoothing_factor": 0.0,
+  "bf16": true,
+  "fp16": false,
+  "bf16_full_eval": false,
+  "fp16_full_eval": false,
+  "tf32": null,
+  "gradient_checkpointing": true,
+  "gradient_checkpointing_kwargs": null,
+  "torch_compile": false,
+  "torch_compile_backend": null,
+  "torch_compile_mode": null,
+  "use_liger_kernel": false,
+  "liger_kernel_config": null,
+  "use_cache": false,
+  "neftune_noise_alpha": null,
+  "torch_empty_cache_steps": null,
+  "auto_find_batch_size": false,
+  "logging_strategy": "steps",
+  "logging_steps": 1,
+  "logging_first_step": true,
+  "log_on_each_node": true,
+  "logging_nan_inf_filter": true,
+  "include_num_input_tokens_seen": false,
+  "log_level": "passive",
+  "log_level_replica": "warning",
+  "disable_tqdm": null,
+  "report_to": [
+    "tensorboard"
+  ],
+  "run_name": "/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_004/solver",
+  "project": "huggingface",
+  "trackio_space_id": null,
+  "trackio_bucket_id": null,
+  "trackio_static_space_id": null,
+  "eval_strategy": "no",
+  "eval_steps": 16.0,
+  "eval_delay": 0,
+  "per_device_eval_batch_size": 1,
+  "prediction_loss_only": false,
+  "eval_on_start": false,
+  "eval_do_concat_batches": true,
+  "eval_use_gather_object": false,
+  "eval_accumulation_steps": null,
+  "include_for_metrics": [],
+  "batch_eval_metrics": false,
+  "save_only_model": false,
+  "save_strategy": "steps",
+  "save_steps": 16.0,
+  "save_on_each_node": false,
+  "save_total_limit": 1,
+  "enable_jit_checkpoint": false,
+  "push_to_hub": false,
+  "hub_token": null,
+  "hub_private_repo": null,
+  "hub_model_id": null,
+  "hub_strategy": "every_save",
+  "hub_always_push": false,
+  "hub_revision": null,
+  "load_best_model_at_end": false,
+  "metric_for_best_model": "reward",
+  "greater_is_better": true,
+  "ignore_data_skip": false,
+  "restore_callback_states_from_checkpoint": false,
+  "full_determinism": false,
+  "seed": 42,
+  "data_seed": 42,
+  "use_cpu": false,
+  "accelerator_config": {
+    "dispatch_batches": false
+  },
+  "parallelism_config": null,
+  "dataloader_drop_last": false,
+  "dataloader_num_workers": null,
+  "dataloader_pin_memory": true,
+  "dataloader_persistent_workers": true,
+  "dataloader_prefetch_factor": null,
+  "dataloader_multiprocessing_context": null,
+  "dataloader_in_order": true,
+  "remove_unused_columns": false,
+  "label_names": null,
+  "train_sampling_strategy": "random",
+  "length_column_name": "length",
+  "ddp_find_unused_parameters": false,
+  "ddp_bucket_cap_mb": null,
+  "ddp_broadcast_buffers": null,
+  "ddp_static_graph": null,
+  "ddp_backend": null,
+  "ddp_timeout": 18000000,
+  "fsdp": [],
+  "fsdp_config": null,
+  "deepspeed": {
+    "fp16": {
+      "enabled": "auto",
+      "loss_scale": 0,
+      "loss_scale_window": 1000,
+      "initial_scale_power": 16,
+      "hysteresis": 2,
+      "min_loss_scale": 1
+    },
+    "bf16": {
+      "enabled": "auto"
+    },
+    "zero_optimization": {
+      "stage": 3,
+      "offload_optimizer": {
+        "device": "none",
+        "pin_memory": true
+      },
+      "offload_param": {
+        "device": "none",
+        "pin_memory": true
+      },
+      "overlap_comm": false,
+      "contiguous_gradients": true,
+      "sub_group_size": 1000000000.0,
+      "reduce_bucket_size": "auto",
+      "zero_quantized_weights": false,
+      "zero_quantized_gradients": false,
+      "stage3_prefetch_bucket_size": "auto",
+      "stage3_param_persistence_threshold": "auto",
+      "stage3_max_live_parameters": 1000000000.0,
+      "stage3_max_reuse_distance": 1000000000.0,
+      "stage3_gather_16bit_weights_on_model_save": true
+    },
+    "gradient_accumulation_steps": "auto",
+    "gradient_clipping": "auto",
+    "steps_per_print": 2000,
+    "train_batch_size": "auto",
+    "train_micro_batch_size_per_gpu": "auto",
+    "wall_clock_breakdown": false
+  },
+  "debug": null,
+  "skip_memory_metrics": true,
+  "do_train": false,
+  "do_eval": false,
+  "do_predict": false,
+  "resume_from_checkpoint": null,
+  "local_rank": 0,
+  "sortish_sampler": false,
+  "predict_with_generate": false,
+  "generation_max_length": null,
+  "generation_num_beams": null,
+  "generation_config": null,
+  "tuner_backend": "peft",
+  "vit_gradient_checkpointing": true,
+  "router_aux_loss_coef": 0.0,
+  "enable_dft_loss": false,
+  "enable_channel_loss": false,
+  "safe_serialization": true,
+  "max_shard_size": "5GB",
+  "check_model": true,
+  "acc_strategy": "token",
+  "train_dataloader_shuffle": true,
+  "group_by_length": false,
+  "max_epochs": null,
+  "aligner_lr": null,
+  "vit_lr": null,
+  "use_logits_to_keep": null,
+  "ds3_gather_for_generation": true,
+  "resume_only_model": false,
+  "optimizer": null,
+  "loss_type": "grpo",
+  "mrl_dims": null,
+  "eval_metric": null,
+  "callbacks": [
+    "cvis_step_timing"
+  ],
+  "early_stop_interval": null,
+  "eval_use_evalscope": false,
+  "eval_dataset": [],
+  "eval_dataset_args": null,
+  "eval_limit": null,
+  "eval_generation_config": null,
+  "extra_eval_args": null,
+  "tuner_type": "full",
+  "use_galore": false,
+  "galore_target_modules": null,
+  "galore_rank": 128,
+  "galore_update_proj_gap": 50,
+  "galore_scale": 1.0,
+  "galore_proj_type": "std",
+  "galore_optim_per_parameter": false,
+  "galore_with_embedding": false,
+  "galore_quantization": false,
+  "galore_proj_quant": false,
+  "galore_proj_bits": 4,
+  "galore_proj_group_size": 256,
+  "galore_cos_threshold": 0.4,
+  "galore_gamma_proj": 2,
+  "galore_queue_size": 5,
+  "lisa_activated_layers": 0,
+  "lisa_step_interval": 20,
+  "use_flash_ckpt": false,
+  "logging_dir": "/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_004/solver/runs",
+  "warmup_ratio": 0.03,
+  "use_ray": false,
+  "ray_exp_name": null,
+  "device_groups": null,
+  "model": "/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_003/solver/checkpoint-160",
+  "model_type": "qwen3_5",
+  "model_revision": null,
+  "task_type": "causal_lm",
+  "torch_dtype": "bfloat16",
+  "attn_impl": null,
+  "experts_impl": null,
+  "new_special_tokens": [],
+  "num_labels": null,
+  "problem_type": null,
+  "rope_scaling": null,
+  "device_map": null,
+  "max_memory": {},
+  "max_model_len": null,
+  "local_repo_path": null,
+  "init_strategy": null,
+  "template": "qwen3_5",
+  "system": "Base your inference and reasoning strictly on the provided image. Provide the final answer in <answer> and </answer>.",
+  "max_length": 2048,
+  "truncation_strategy": "left",
+  "max_pixels": null,
+  "agent_template": null,
+  "norm_bbox": null,
+  "use_chat_template": true,
+  "padding_side": "right",
+  "padding_free": false,
+  "loss_scale": "last_round+ignore_empty_think",
+  "sequence_parallel_size": 1,
+  "is_binary_loss_scale": null,
+  "template_backend": "swift",
+  "response_prefix": null,
+  "enable_thinking": false,
+  "preserve_thinking": null,
+  "add_non_thinking_prefix": true,
+  "disable_ignore_empty_think": false,
+  "dataset": [
+    "/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_004/solver/train.jsonl"
+  ],
+  "val_dataset": [],
+  "cached_dataset": [],
+  "cached_val_dataset": [],
+  "split_dataset_ratio": 0.0,
+  "dataset_num_proc": 1,
+  "load_from_cache_file": false,
+  "dataset_shuffle": true,
+  "val_dataset_shuffle": false,
+  "streaming": false,
+  "interleave_prob": null,
+  "stopping_strategy": "first_exhausted",
+  "shuffle_buffer_size": 1000,
+  "download_mode": "reuse_dataset_if_exists",
+  "columns": {},
+  "strict": false,
+  "disable_auto_column_mapping": false,
+  "model_name": null,
+  "model_author": null,
+  "custom_dataset_info": [],
+  "decision_drop_instructions": 0.0,
+  "decision_subsample_options": 0,
+  "decision_state_as_text": 0.0,
+  "decision_shuffle_fields": false,
+  "quant_method": null,
+  "quant_bits": null,
+  "hqq_axis": null,
+  "bnb_4bit_compute_dtype": "bfloat16",
+  "bnb_4bit_quant_type": "nf4",
+  "bnb_4bit_use_double_quant": true,
+  "bnb_4bit_quant_storage": null,
+  "max_new_tokens": 2048,
+  "temperature": 1.0,
+  "top_k": -1,
+  "top_p": 1.0,
+  "repetition_penalty": 1.0,
+  "num_beams": 1,
+  "stream": false,
+  "stop_words": [],
+  "logprobs": false,
+  "top_logprobs": null,
+  "structured_outputs_regex": null,
+  "adapters": [],
+  "external_plugins": [
+    "/data1/xiangding/cvis-ablate/repo/cvis/plugins/swift_reward.py"
+  ],
+  "custom_register_path": [],
+  "model_kwargs": {},
+  "enable_npu_model_patch": true,
+  "load_args": false,
+  "load_data_args": false,
+  "packing": false,
+  "packing_length": null,
+  "packing_num_proc": 1,
+  "packing_strategy": "binpack",
+  "lazy_tokenize": true,
+  "use_hf": false,
+  "ignore_args_error": false,
+  "use_swift_lora": false,
+  "freeze_parameters": [],
+  "freeze_parameters_regex": null,
+  "freeze_parameters_ratio": 0.0,
+  "trainable_parameters": [
+    "model.visual.merger"
+  ],
+  "trainable_parameters_regex": null,
+  "freeze_llm": false,
+  "freeze_vit": false,
+  "freeze_aligner": false,
+  "target_modules": [
+    "all-linear"
+  ],
+  "target_regex": null,
+  "target_parameters": null,
+  "modules_to_save": [],
+  "lora_rank": 8,
+  "lora_alpha": 32.0,
+  "lora_dropout": 0.05,
+  "lora_bias": "none",
+  "lora_dtype": null,
+  "lorap_lr_ratio": null,
+  "use_rslora": false,
+  "use_dora": false,
+  "lora_ga_batch_size": 2,
+  "lora_ga_iters": 2,
+  "lora_ga_max_length": 1024,
+  "lora_ga_direction": "ArB2r",
+  "lora_ga_scale": "stable",
+  "lora_ga_stable_gamma": 16,
+  "init_weights": true,
+  "fourier_n_frequency": 2000,
+  "fourier_scaling": 300.0,
+  "boft_block_size": 4,
+  "boft_block_num": 0,
+  "boft_n_butterfly_factor": 1,
+  "boft_dropout": 0.0,
+  "vera_rank": 256,
+  "vera_projection_prng_key": 0,
+  "vera_dropout": 0.0,
+  "vera_d_initial": 0.1,
+  "adapter_act": "gelu",
+  "adapter_length": 128,
+  "adalora_target_r": 8,
+  "adalora_init_r": 12,
+  "adalora_tinit": 0,
+  "adalora_tfinal": 0,
+  "adalora_deltaT": 1,
+  "adalora_beta1": 0.85,
+  "adalora_beta2": 0.85,
+  "adalora_orth_reg_weight": 0.5,
+  "llamapro_num_new_blocks": 4,
+  "llamapro_num_groups": null,
+  "reft_layer_key": null,
+  "reft_layers": null,
+  "reft_rank": 4,
+  "reft_intervention_type": "LoreftIntervention",
+  "reft_args": null,
+  "swanlab_token": null,
+  "swanlab_project": "ms-swift",
+  "swanlab_workspace": null,
+  "swanlab_exp_name": null,
+  "swanlab_notification_method": null,
+  "swanlab_webhook_url": null,
+  "swanlab_secret": null,
+  "swanlab_sender_email": null,
+  "swanlab_receiver_email": null,
+  "swanlab_smtp_server": null,
+  "swanlab_smtp_port": null,
+  "swanlab_email_language": "zh",
+  "swanlab_mode": "cloud",
+  "add_version": false,
+  "create_checkpoint_symlink": false,
+  "zero_hpz_partition_size": null,
+  "deepspeed_autotp_size": null,
+  "reward_model": null,
+  "reward_adapters": [],
+  "reward_model_type": null,
+  "reward_model_revision": null,
+  "reward_template": null,
+  "num_ppo_epochs": 4,
+  "whiten_rewards": false,
+  "kl_coef": 0.05,
+  "cliprange": 0.2,
+  "vf_coef": 0.1,
+  "cliprange_value": 0.2,
+  "gamma": 1.0,
+  "lam": 0.95,
+  "num_mini_batches": 1,
+  "local_rollout_forward_batch_size": 64,
+  "num_sample_generations": 10,
+  "response_length": 2048,
+  "missing_eos_penalty": null,
+  "vllm_gpu_memory_utilization": 0.2,
+  "vllm_tensor_parallel_size": 1,
+  "vllm_pipeline_parallel_size": 1,
+  "vllm_enable_expert_parallel": false,
+  "vllm_max_num_seqs": null,
+  "vllm_max_model_len": 4096,
+  "vllm_disable_custom_all_reduce": true,
+  "vllm_enforce_eager": false,
+  "vllm_limit_mm_per_prompt": null,
+  "vllm_max_lora_rank": 16,
+  "vllm_enable_prefix_caching": true,
+  "vllm_use_async_engine": null,
+  "vllm_quantization": null,
+  "vllm_reasoning_parser": null,
+  "vllm_disable_cascade_attn": false,
+  "vllm_mm_processor_cache_gb": null,
+  "vllm_speculative_config": null,
+  "vllm_engine_kwargs": {},
+  "vllm_data_parallel_size": 1,
+  "min_p": 0.0,
+  "use_vllm": true,
+  "vllm_mode": "colocate",
+  "vllm_enable_lora": false,
+  "vllm_server_base_url": null,
+  "vllm_server_host": null,
+  "vllm_server_port": [
+    8000
+  ],
+  "vllm_server_timeout": 240.0,
+  "vllm_server_group_port": null,
+  "enable_flattened_weight_sync": true,
+  "async_generate": false,
+  "sleep_level": 0,
+  "move_model_batches": null,
+  "offload_optimizer": false,
+  "offload_model": false,
+  "wandb_log_unique_prompts": null,
+  "generation_batch_size": 256,
+  "steps_per_generation": null,
+  "teacher_tag_key": "dataset",
+  "epsilon": 0.2,
+  "epsilon_high": null,
+  "delta": null,
+  "cosine_min_len_value_wrong": -0.5,
+  "cosine_max_len_value_wrong": 0.0,
+  "cosine_min_len_value_correct": 1.0,
+  "cosine_max_len_value_correct": 0.5,
+  "cosine_max_len": null,
+  "repetition_n_grams": 3,
+  "repetition_max_penalty": -1.0,
+  "reward_model_plugin": null,
+  "chord_sft_dataset": [],
+  "chord_sft_per_device_train_batch_size": null,
+  "chord_enable_phi_function": false,
+  "chord_mu_warmup_steps": null,
+  "chord_mu_decay_steps": null,
+  "chord_mu_peak": null,
+  "chord_mu_valley": null,
+  "sync_ref_model": false,
+  "ref_model_sync_steps": 512,
+  "ref_model_mixup_alpha": 0.6,
+  "multi_turn_scheduler": null,
+  "max_turns": null,
+  "completion_length_limit_scope": "per_round",
+  "vllm_server_pass_dataset": false,
+  "use_gym_env": null,
+  "gym_env": null,
+  "dynamic_sample": false,
+  "max_resample_times": 3,
+  "overlong_filter": false,
+  "soft_max_length": null,
+  "soft_cache_length": null,
+  "scale_rewards": "group",
+  "log_entropy": false,
+  "top_entropy_quantile": 1.0,
+  "importance_sampling_level": "token",
+  "tau_pos": 1.0,
+  "tau_neg": 1.05,
+  "advantage_estimator": "grpo",
+  "kl_in_reward": false,
+  "teacher_kl_coef": 1.0,
+  "advantage_reweight": null,
+  "rlsd_lambda": 0.5,
+  "rlsd_reweight_clip_range": 0.2,
+  "rlsd_lambda_warmup_steps": 0,
+  "rlsd_lambda_decay_steps": 0,
+  "rlsd_negative_only": false,
+  "sdar_loss_coef": 0.0,
+  "sdar_gate_beta": 5.0,
+  "real_tau": 0.5,
+  "fipo_decay_rate": 32.0,
+  "fipo_clip_range": 0.2,
+  "fipo_clip_high_only": true,
+  "fipo_safety_threshold": 4.0,
+  "num_generations_eval": null,
+  "rollout_importance_sampling_mode": null,
+  "rollout_importance_sampling_threshold": 2.0,
+  "log_rollout_offpolicy_metrics": false,
+  "off_policy_sequence_mask_delta": null,
+  "num_generations": 8,
+  "reward_funcs": [
+    "cvis_majority_accuracy"
+  ],
+  "reward_weights": null,
+  "log_completions": false,
+  "num_iterations": 1,
+  "teacher_model": null,
+  "teacher_adapters": [],
+  "teacher_model_type": null,
+  "teacher_model_revision": null,
+  "teacher_deepspeed": null,
+  "teacher_model_server": null,
+  "offload_teacher_model": false,
+  "rlhf_type": "grpo",
+  "ref_model": "/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_003/solver/checkpoint-160",
+  "ref_adapters": [],
+  "ref_model_type": "qwen3_5",
+  "ref_model_revision": null,
+  "beta": 0.01,
+  "label_smoothing": 0,
+  "max_completion_length": 2048,
+  "rpo_alpha": null,
+  "ld_alpha": null,
+  "discopop_tau": 0.05,
+  "loss_weights": null,
+  "cpo_alpha": 1.0,
+  "simpo_gamma": 1,
+  "desirable_weight": 1.0,
+  "undesirable_weight": 1.0,
+  "center_rewards_coefficient": null,
+  "sft_alpha": 0,
+  "lmbda": 0.5,
+  "seq_kd": false,
+  "gkd_logits_topk": null,
+  "swift_version": "4.6.0",
+  "ckpt_dir": "/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_003/solver/checkpoint-160",
+  "rank": 0,
+  "global_world_size": 4,
+  "local_world_size": 4,
+  "model_suffix": "checkpoint-160",
+  "model_info": "ModelInfo(model_type='qwen3_5', model_dir='/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_003/solver/checkpoint-160', torch_dtype=torch.bfloat16, max_model_len=262144, quant_method=None, quant_bits=None, rope_scaling=None, is_moe_model=False, is_multimodal=True, config=None, task_type='causal_lm', num_labels=None)",
+  "model_meta": "ModelMeta(model_type='qwen3_5', model_groups=[ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3.5-0.8B', hf_model_id='Qwen/Qwen3.5-0.8B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-2B', hf_model_id='Qwen/Qwen3.5-2B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-4B', hf_model_id='Qwen/Qwen3.5-4B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-9B', hf_model_id='Qwen/Qwen3.5-9B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-27B', hf_model_id='Qwen/Qwen3.5-27B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-27B-FP8', hf_model_id='Qwen/Qwen3.5-27B-FP8', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-0.8B-Base', hf_model_id='Qwen/Qwen3.5-0.8B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-2B-Base', hf_model_id='Qwen/Qwen3.5-2B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-4B-Base', hf_model_id='Qwen/Qwen3.5-4B-Base', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.5-9B-Base', hf_model_id='Qwen/Qwen3.5-9B-Base', model_path=None, ms_revision=None, hf_revision=None)], template='qwen3_5', ignore_patterns=None, requires=None, tags=[]), ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3.6-27B', hf_model_id='Qwen/Qwen3.6-27B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.6-27B-FP8', hf_model_id='Qwen/Qwen3.6-27B-FP8', model_path=None, ms_revision=None, hf_revision=None)], template='qwen3_5', ignore_patterns=None, requires=None, tags=[]), ModelGroup(models=[Model(ms_model_id='Qwen/Qwen3.8-27B', hf_model_id='Qwen/Qwen3.8-27B', model_path=None, ms_revision=None, hf_revision=None), Model(ms_model_id='Qwen/Qwen3.8-27B-FP8', hf_model_id='Qwen/Qwen3.8-27B-FP8', model_path=None, ms_revision=None, hf_revision=None)], template='qwen3_8', ignore_patterns=None, requires=None, tags=[])], loader=<class 'swift.model.models.qwen.Qwen3_5Loader'>, template=None, model_arch=MultiModelKeys(arch_name='qwen2_vl', embedding=None, module_list=None, lm_head=None, q_proj=None, k_proj=None, v_proj=None, o_proj=None, attention=None, mlp='model.language_model.layers.{}.mlp', down_proj=None, qkv_proj=None, qk_proj=None, qa_proj=None, qb_proj=None, kv_proj=None, kva_proj=None, kvb_proj=None, language_model=['model.language_model', 'lm_head'], aligner=['model.visual.merger'], vision_tower=['model.visual'], generator=[]), mcore_model_type=None, architectures=['Qwen3_5ForConditionalGeneration'], additional_saved_files=[], torch_dtype=None, is_multimodal=True, is_reward=False, task_type=None, ignore_patterns=None, requires=['transformers>=5.2.0', 'qwen_vl_utils>=0.0.14', 'decord'], tags=['vision', 'video'])",
+  "model_dir": "/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_003/solver/checkpoint-160",
+  "template_meta": "QwenTemplateMeta(template_type='qwen3_5', prefix=[], prompt=['<|im_start|>user\\n{{QUERY}}<|im_end|>\\n<|im_start|>assistant\\n'], chat_sep=['<|im_end|>\\n'], suffix=['<|im_end|>\\n'], template_cls=<class 'swift.template.templates.qwen.Qwen3_5Template'>, system_prefix=['<|im_start|>system\\n{{SYSTEM}}<|im_end|>\\n'], default_system=None, auto_add_bos=False, stop_words=['<|endoftext|>'], agent_template='qwen3_5', is_thinking=True, thinking_prefix='<think>\\n', non_thinking_prefix='<think>\\n\\n</think>\\n\\n', preserve_thinking=None, history_thinking_prefix='')",
+  "_val_dataset_exists": false,
+  "hub": "<class 'swift.hub.hub.MSHub'>",
+  "evaluation_strategy": "steps",
+  "training_args": "GRPOConfig(output_dir='/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_004/solver', per_device_train_batch_size=1, num_train_epochs=3.0, max_steps=160, learning_rate=1e-06, lr_scheduler_type=<SchedulerType.COSINE: 'cosine'>, lr_scheduler_kwargs=None, warmup_steps=0.03, optim=<OptimizerNames.ADAMW_TORCH_FUSED: 'adamw_torch_fused'>, optim_args=None, weight_decay=0.1, adam_beta1=0.9, adam_beta2=0.95, adam_epsilon=1e-08, optim_target_modules=None, gradient_accumulation_steps=8, average_tokens_across_devices=None, max_grad_norm=1.0, label_smoothing_factor=0.0, bf16=True, fp16=False, bf16_full_eval=False, fp16_full_eval=False, tf32=None, gradient_checkpointing=True, gradient_checkpointing_kwargs=None, torch_compile=False, torch_compile_backend=None, torch_compile_mode=None, use_liger_kernel=False, liger_kernel_config=None, use_cache=False, neftune_noise_alpha=None, torch_empty_cache_steps=None, auto_find_batch_size=False, logging_strategy=<IntervalStrategy.STEPS: 'steps'>, logging_steps=1, logging_first_step=True, log_on_each_node=True, logging_nan_inf_filter=True, include_num_input_tokens_seen=None, log_level='passive', log_level_replica='warning', disable_tqdm=False, report_to=['tensorboard'], run_name='/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_004/solver', project='huggingface', trackio_space_id=None, trackio_bucket_id=None, trackio_static_space_id=None, eval_strategy=<IntervalStrategy.NO: 'no'>, eval_steps=16.0, eval_delay=0, per_device_eval_batch_size=1, prediction_loss_only=False, eval_on_start=False, eval_do_concat_batches=True, eval_use_gather_object=False, eval_accumulation_steps=None, include_for_metrics=[], batch_eval_metrics=False, save_only_model=False, save_strategy=<SaveStrategy.STEPS: 'steps'>, save_steps=16, save_on_each_node=False, save_total_limit=1, enable_jit_checkpoint=False, push_to_hub=False, hub_token=None, hub_private_repo=None, hub_model_id=None, hub_strategy=<HubStrategy.EVERY_SAVE: 'every_save'>, hub_always_push=False, hub_revision=None, load_best_model_at_end=False, metric_for_best_model='reward', greater_is_better=True, ignore_data_skip=False, restore_callback_states_from_checkpoint=False, full_determinism=False, seed=42, data_seed=42, use_cpu=False, accelerator_config=AcceleratorConfig(split_batches=False, dispatch_batches=False, even_batches=True, use_seedable_sampler=True, non_blocking=False, gradient_accumulation_kwargs=None, use_configured_state=False), parallelism_config=None, dataloader_drop_last=True, dataloader_num_workers=1, dataloader_pin_memory=True, dataloader_persistent_workers=True, dataloader_prefetch_factor=2, dataloader_multiprocessing_context=None, dataloader_in_order=True, remove_unused_columns=False, label_names=None, train_sampling_strategy='random', length_column_name='length', ddp_find_unused_parameters=False, ddp_bucket_cap_mb=None, ddp_broadcast_buffers=None, ddp_static_graph=None, ddp_backend=None, ddp_timeout=18000000, fsdp=[], fsdp_config=None, deepspeed={'fp16': {'enabled': 'auto', 'loss_scale': 0, 'loss_scale_window': 1000, 'initial_scale_power': 16, 'hysteresis': 2, 'min_loss_scale': 1}, 'bf16': {'enabled': 'auto'}, 'zero_optimization': {'stage': 3, 'offload_optimizer': {'device': 'none', 'pin_memory': True}, 'offload_param': {'device': 'none', 'pin_memory': True}, 'overlap_comm': False, 'contiguous_gradients': True, 'sub_group_size': 1000000000.0, 'reduce_bucket_size': 'auto', 'zero_quantized_weights': False, 'zero_quantized_gradients': False, 'stage3_prefetch_bucket_size': 0, 'stage3_param_persistence_threshold': 'auto', 'stage3_max_live_parameters': 1000000000.0, 'stage3_max_reuse_distance': 1000000000.0, 'stage3_gather_16bit_weights_on_model_save': True}, 'gradient_accumulation_steps': 'auto', 'gradient_clipping': 'auto', 'steps_per_print': 2000, 'train_batch_size': 'auto', 'train_micro_batch_size_per_gpu': 'auto', 'wall_clock_breakdown': False}, debug=[], skip_memory_metrics=True, do_train=False, do_eval=False, do_predict=False, resume_from_checkpoint=None, local_rank=0, model_init_kwargs=None, disable_dropout=False, cast_lm_head_to_fp32=False, num_generations=8, num_generations_eval=None, max_completion_length=2048, ds3_gather_for_generation=True, shuffle_dataset=True, generation_batch_size=256, steps_per_generation=64, temperature=1.0, top_p=1.0, top_k=-1, min_p=0.0, generation_kwargs=None, chat_template_kwargs=None, repetition_penalty=1.0, use_transformers_paged=False, cache_implementation=None, use_vllm=True, vllm_mode='colocate', vllm_model_impl='vllm', vllm_enable_sleep_mode=False, vllm_structured_outputs_regex=None, vllm_server_base_url=None, vllm_server_host=None, vllm_server_port=[8000], vllm_server_timeout=240.0, vllm_group_port=51216, vllm_gpu_memory_utilization=0.2, vllm_max_model_length=None, vllm_tensor_parallel_size=1, beta=0.01, num_iterations=1, epsilon=0.2, delta=None, epsilon_high=None, sapo_temperature_neg=1.05, sapo_temperature_pos=1.0, importance_sampling_level='token', reward_weights=None, multi_objective_aggregation='sum_then_normalize', scale_rewards='group', loss_type='grpo', mask_truncated_completions=False, sync_ref_model=False, ref_model_mixup_alpha=0.6, ref_model_sync_steps=512, top_entropy_quantile=1.0, max_tool_calling_iterations=None, vllm_importance_sampling_correction=True, vllm_importance_sampling_mode='sequence_mask', vllm_importance_sampling_cap=3.0, off_policy_mask_threshold=None, use_bias_correction_kl=False, log_completions=False, num_completions_to_print=None, log_unique_prompts=False, log_completions_hub_repo=None, tuner_backend='peft', vit_gradient_checkpointing=True, router_aux_loss_coef=0.0, enable_dft_loss=False, enable_channel_loss=False, safe_serialization=True, max_shard_size='5GB', check_model=True, acc_strategy='token', train_dataloader_shuffle=True, group_by_length=False, max_epochs=None, aligner_lr=None, vit_lr=None, use_logits_to_keep=None, resume_only_model=False, optimizer=None, mrl_dims=None, eval_metric=None, callbacks=['cvis_step_timing'], early_stop_interval=None, eval_use_evalscope=False, eval_dataset=[], eval_dataset_args=None, eval_limit=None, eval_generation_config=None, extra_eval_args=None, tuner_type='full', use_galore=False, galore_target_modules=None, galore_rank=128, galore_update_proj_gap=50, galore_scale=1.0, galore_proj_type='std', galore_optim_per_parameter=False, galore_with_embedding=False, galore_quantization=False, galore_proj_quant=False, galore_proj_bits=4, galore_proj_group_size=256, galore_cos_threshold=0.4, galore_gamma_proj=2, galore_queue_size=5, lisa_activated_layers=0, lisa_step_interval=20, use_flash_ckpt=False, logging_dir='/data1/xiangding/cvis-ablate/runs/a6_coupled_proposer_coder/run_a6_coupled_proposer_coder_4g/iteration_004/solver/runs', warmup_ratio=0.0, vllm_pipeline_parallel_size=1, vllm_enable_expert_parallel=False, vllm_max_num_seqs=None, vllm_max_model_len=4096, vllm_disable_custom_all_reduce=True, vllm_enforce_eager=False, vllm_limit_mm_per_prompt=None, vllm_max_lora_rank=16, vllm_enable_prefix_caching=True, vllm_use_async_engine=None, vllm_quantization=None, vllm_reasoning_parser=None, vllm_disable_cascade_attn=False, vllm_mm_processor_cache_gb=None, vllm_speculative_config=None, vllm_engine_kwargs={}, vllm_data_parallel_size=1, stop_words=[], vllm_enable_lora=False, lora_rank=8, vllm_server_group_port=None, enable_flattened_weight_sync=True, async_generate=False, structured_outputs_regex=None, sleep_level=0, move_model_batches=None, offload_optimizer=False, offload_model=False, wandb_log_unique_prompts=None, teacher_tag_key='dataset', cosine_min_len_value_wrong=-0.5, cosine_max_len_value_wrong=0.0, cosine_min_len_value_correct=1.0, cosine_max_len_value_correct=0.5, cosine_max_len=2048, repetition_n_grams=3, repetition_max_penalty=-1.0, reward_model=None, reward_model_plugin=None, chord_sft_dataset=[], chord_sft_per_device_train_batch_size=None, chord_enable_phi_function=False, chord_mu_warmup_steps=None, chord_mu_decay_steps=None, chord_mu_peak=None, chord_mu_valley=None, multi_turn_scheduler=None, max_turns=None, completion_length_limit_scope='per_round', vllm_server_pass_dataset=False, use_gym_env=None, gym_env=None, dynamic_sample=False, max_resample_times=3, overlong_filter=False, soft_max_length=None, soft_cache_length=None, log_entropy=False, tau_pos=1.0, tau_neg=1.05, advantage_estimator='grpo', kl_in_reward=False, teacher_kl_coef=1.0, advantage_reweight=None, rlsd_lambda=0.5, rlsd_reweight_clip_range=0.2, rlsd_lambda_warmup_steps=0, rlsd_lambda_decay_steps=0, rlsd_negative_only=False, sdar_loss_coef=0.0, sdar_gate_beta=5.0, real_tau=0.5, fipo_decay_rate=32.0, fipo_clip_range=0.2, fipo_clip_high_only=True, fipo_safety_threshold=4.0, dataset_shuffle=True, rollout_importance_sampling_mode=None, rollout_importance_sampling_threshold=2.0, log_rollout_offpolicy_metrics=False, off_policy_sequence_mask_delta=None, offload_teacher_model=False)",
+  "_teacher_use_disable_adapter": false
+}

a6_coupled_proposer_coder/round_4/chat_template.jinja ADDED Viewed

	@@ -0,0 +1,154 @@

+{%- set image_count = namespace(value=0) %}
+{%- set video_count = namespace(value=0) %}
+{%- macro render_content(content, do_vision_count, is_system_content=false) %}
+    {%- if content is string %}
+        {{- content }}
+    {%- elif content is iterable and content is not mapping %}
+        {%- for item in content %}
+            {%- if 'image' in item or 'image_url' in item or item.type == 'image' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain images.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set image_count.value = image_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Picture ' ~ image_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|image_pad|><|vision_end|>' }}
+            {%- elif 'video' in item or item.type == 'video' %}
+                {%- if is_system_content %}
+                    {{- raise_exception('System message cannot contain videos.') }}
+                {%- endif %}
+                {%- if do_vision_count %}
+                    {%- set video_count.value = video_count.value + 1 %}
+                {%- endif %}
+                {%- if add_vision_id %}
+                    {{- 'Video ' ~ video_count.value ~ ': ' }}
+                {%- endif %}
+                {{- '<|vision_start|><|video_pad|><|vision_end|>' }}
+            {%- elif 'text' in item %}
+                {{- item.text }}
+            {%- else %}
+                {{- raise_exception('Unexpected item type in content.') }}
+            {%- endif %}
+        {%- endfor %}
+    {%- elif content is none or content is undefined %}
+        {{- '' }}
+    {%- else %}
+        {{- raise_exception('Unexpected content type.') }}
+    {%- endif %}
+{%- endmacro %}
+{%- if not messages %}
+    {{- raise_exception('No messages provided.') }}
+{%- endif %}
+{%- if tools and tools is iterable and tools is not mapping %}
+    {{- '<|im_start|>system\n' }}
+    {{- "# Tools\n\nYou have access to the following functions:\n\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>" }}
+    {{- '\n\nIf you choose to call a function ONLY reply in the following format with NO suffix:\n\n<tool_call>\n<function=example_function_name>\n<parameter=example_parameter_1>\nvalue_1\n</parameter>\n<parameter=example_parameter_2>\nThis is the value for the second parameter\nthat can span\nmultiple lines\n</parameter>\n</function>\n</tool_call>\n\n<IMPORTANT>\nReminder:\n- Function calls MUST follow the specified format: an inner <function=...></function> block must be nested within <tool_call></tool_call> XML tags\n- Required parameters MUST be specified\n- You may provide optional reasoning for your function call in natural language BEFORE the function call, but NOT after\n- If there is no function call available, answer the question like normal with your current knowledge and do not tell the user about function calls\n</IMPORTANT>' }}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {%- if content %}
+            {{- '\n\n' + content }}
+        {%- endif %}
+    {%- endif %}
+    {{- '<|im_end|>\n' }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {%- set content = render_content(messages[0].content, false, true)|trim %}
+        {{- '<|im_start|>system\n' + content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" %}
+        {%- set content = render_content(message.content, false)|trim %}
+        {%- if not(content.startswith('<tool_response>') and content.endswith('</tool_response>')) %}
+            {%- set ns.multi_step_tool = false %}
+            {%- set ns.last_query_index = index %}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if ns.multi_step_tool %}
+    {{- raise_exception('No user query found in messages.') }}
+{%- endif %}
+{%- for message in messages %}
+    {%- set content = render_content(message.content, true)|trim %}
+    {%- if message.role == "system" %}
+        {%- if not loop.first %}
+            {{- raise_exception('System message must be at the beginning.') }}
+        {%- endif %}
+    {%- elif message.role == "user" %}
+        {{- '<|im_start|>' + message.role + '\n' + content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is string %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in content %}
+                {%- set reasoning_content = content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+                {%- set content = content.split('</think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- set reasoning_content = reasoning_content|trim %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content + '\n</think>\n\n' + content }}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls and message.tool_calls is iterable and message.tool_calls is not mapping %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if tool_call.function is defined %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {%- if loop.first %}
+                    {%- if content|trim %}
+                        {{- '\n\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- else %}
+                        {{- '<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                    {%- endif %}
+                {%- else %}
+                    {{- '\n<tool_call>\n<function=' + tool_call.name + '>\n' }}
+                {%- endif %}
+                {%- if tool_call.arguments is defined %}
+                    {%- for args_name, args_value in tool_call.arguments|items %}
+                        {{- '<parameter=' + args_name + '>\n' }}
+                        {%- set args_value = args_value | tojson | safe if args_value is mapping or (args_value is sequence and args_value is not string) else args_value | string %}
+                        {{- args_value }}
+                        {{- '\n</parameter>\n' }}
+                    {%- endfor %}
+                {%- endif %}
+                {{- '</function>\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.previtem and loop.previtem.role != "tool" %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- content }}
+        {{- '\n</tool_response>' }}
+        {%- if not loop.last and loop.nextitem.role != "tool" %}
+            {{- '<|im_end|>\n' }}
+        {%- elif loop.last %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- else %}
+        {{- raise_exception('Unexpected message role.') }}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- else %}
+        {{- '<think>\n' }}
+    {%- endif %}
+{%- endif %}

a6_coupled_proposer_coder/round_4/config.json ADDED Viewed

	@@ -0,0 +1,117 @@

+{
+  "architectures": [
+    "Qwen3_5ForConditionalGeneration"
+  ],
+  "dtype": "bfloat16",
+  "eos_token_id": 248046,
+  "hidden_size": 4096,
+  "image_token_id": 248056,
+  "model_type": "qwen3_5",
+  "pad_token_id": 248044,
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "attn_output_gate": true,
+    "bos_token_id": null,
+    "dtype": "bfloat16",
+    "eos_token_id": 248044,
+    "full_attention_interval": 4,
+    "head_dim": 256,
+    "hidden_act": "silu",
+    "hidden_size": 4096,
+    "initializer_range": 0.02,
+    "intermediate_size": 12288,
+    "layer_types": [
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention",
+      "linear_attention",
+      "linear_attention",
+      "linear_attention",
+      "full_attention"
+    ],
+    "linear_conv_kernel_dim": 4,
+    "linear_key_head_dim": 128,
+    "linear_num_key_heads": 16,
+    "linear_num_value_heads": 32,
+    "linear_value_head_dim": 128,
+    "mamba_ssm_dtype": "float32",
+    "max_position_embeddings": 262144,
+    "mlp_only_layers": [],
+    "model_type": "qwen3_5_text",
+    "mtp_num_hidden_layers": 1,
+    "mtp_use_dedicated_embeddings": false,
+    "num_attention_heads": 16,
+    "num_hidden_layers": 32,
+    "num_key_value_heads": 4,
+    "pad_token_id": 248044,
+    "partial_rotary_factor": 0.25,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "mrope_interleaved": true,
+      "mrope_section": [
+        11,
+        11,
+        10
+      ],
+      "partial_rotary_factor": 0.25,
+      "rope_theta": 10000000,
+      "rope_type": "default"
+    },
+    "tie_word_embeddings": false,
+    "use_cache": false,
+    "vocab_size": 248320
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "5.17.0",
+  "use_cache": false,
+  "video_token_id": 248057,
+  "vision_config": {
+    "deepstack_visual_indexes": [],
+    "depth": 27,
+    "dtype": "bfloat16",
+    "hidden_act": "gelu_pytorch_tanh",
+    "hidden_size": 1152,
+    "in_channels": 3,
+    "initializer_range": 0.02,
+    "intermediate_size": 4304,
+    "model_type": "qwen3_5_vision",
+    "num_heads": 16,
+    "num_position_embeddings": 2304,
+    "out_hidden_size": 4096,
+    "patch_size": 16,
+    "rope_parameters": {
+      "rope_theta": 10000.0,
+      "rope_type": "axial"
+    },
+    "spatial_merge_size": 2,
+    "temporal_patch_size": 2
+  },
+  "vision_end_token_id": 248054,
+  "vision_start_token_id": 248053
+}

a6_coupled_proposer_coder/round_4/generation_config.json ADDED Viewed

	@@ -0,0 +1,9 @@

+{
+  "_from_model_config": true,
+  "eos_token_id": [
+    248044,
+    248046
+  ],
+  "transformers_version": "5.17.0",
+  "use_cache": true
+}

a6_coupled_proposer_coder/round_4/model-00001-of-00004.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:b40002f4237ac5d8b2a6afb07bdf2f59c902b89fb10173c24310fa34c547d25c
+size 4942706120

a6_coupled_proposer_coder/round_4/model-00002-of-00004.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:0bdc7b3e0b33ea4aba11bfa698c5a16107cdac0fe5d69731c1718c3190ad1031
+size 4987757928

a6_coupled_proposer_coder/round_4/model-00003-of-00004.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:861863660c58d5ac9cf2f08fb0c8075c1867d92bf42e636575c5fdb892bfcabf
+size 4954810240

a6_coupled_proposer_coder/round_4/model-00004-of-00004.safetensors ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:e7d75934e074358e372ec5058add28fd271db893e2bc85e05ed9ba950896c071
+size 3934446832

a6_coupled_proposer_coder/round_4/model.safetensors.index.json ADDED Viewed

	@@ -0,0 +1,768 @@

+{
+  "metadata": {
+    "total_parameters": 1469680,
+    "total_size": 18819627488
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.embed_tokens.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.linear_attn.A_log": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.0.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.linear_attn.A_log": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.linear_attn.in_proj_qkv.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.linear_attn.in_proj_z.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.linear_attn.norm.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.linear_attn.out_proj.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.mlp.down_proj.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.mlp.gate_proj.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.mlp.up_proj.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.1.post_attention_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.10.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.linear_attn.A_log": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.10.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.11.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.linear_attn.A_log": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.12.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.linear_attn.A_log": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.13.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.13.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.13.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.linear_attn.A_log": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.14.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.15.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.linear_attn.A_log": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.16.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.linear_attn.A_log": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.17.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.linear_attn.A_log": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.18.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.19.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.2.input_layernorm.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.2.linear_attn.A_log": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.2.linear_attn.conv1d.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.2.linear_attn.dt_bias": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.2.linear_attn.in_proj_a.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.2.linear_attn.in_proj_b.weight": "model-00001-of-00004.safetensors",
+    "model.language_model.layers.2.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.2.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.2.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.2.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.2.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.2.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.2.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.2.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.20.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.linear_attn.A_log": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.20.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.linear_attn.A_log": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.21.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.linear_attn.A_log": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.22.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.self_attn.k_norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.self_attn.k_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.self_attn.o_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.self_attn.q_norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.self_attn.q_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.23.self_attn.v_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.linear_attn.A_log": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.linear_attn.in_proj_qkv.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.linear_attn.in_proj_z.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.linear_attn.norm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.linear_attn.out_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.mlp.down_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.mlp.gate_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.mlp.up_proj.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.24.post_attention_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.25.input_layernorm.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.25.linear_attn.A_log": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.25.linear_attn.conv1d.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.25.linear_attn.dt_bias": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.25.linear_attn.in_proj_a.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.25.linear_attn.in_proj_b.weight": "model-00003-of-00004.safetensors",
+    "model.language_model.layers.25.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.25.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.25.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.25.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.25.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.25.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.25.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.25.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.linear_attn.A_log": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.26.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.27.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.linear_attn.A_log": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.28.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.linear_attn.A_log": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.29.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.3.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.3.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.3.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.3.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.3.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.3.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.3.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.3.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.3.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.3.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.3.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.30.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.linear_attn.A_log": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.linear_attn.conv1d.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.linear_attn.dt_bias": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.linear_attn.in_proj_a.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.linear_attn.in_proj_b.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.linear_attn.in_proj_qkv.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.linear_attn.in_proj_z.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.linear_attn.norm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.linear_attn.out_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.30.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.input_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.mlp.down_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.mlp.gate_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.mlp.up_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.post_attention_layernorm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.self_attn.k_norm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.self_attn.k_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.self_attn.o_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.self_attn.q_norm.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.self_attn.q_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.31.self_attn.v_proj.weight": "model-00004-of-00004.safetensors",
+    "model.language_model.layers.4.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.linear_attn.A_log": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.4.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.linear_attn.A_log": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.5.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.linear_attn.A_log": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.6.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.self_attn.k_norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.self_attn.k_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.self_attn.o_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.self_attn.q_norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.self_attn.q_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.7.self_attn.v_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.linear_attn.A_log": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.8.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.input_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.linear_attn.A_log": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.linear_attn.conv1d.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.linear_attn.dt_bias": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.linear_attn.in_proj_a.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.linear_attn.in_proj_b.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.linear_attn.in_proj_qkv.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.linear_attn.in_proj_z.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.linear_attn.norm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.linear_attn.out_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.mlp.down_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.mlp.gate_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.mlp.up_proj.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.layers.9.post_attention_layernorm.weight": "model-00002-of-00004.safetensors",
+    "model.language_model.norm.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.0.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.1.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.10.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.11.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.12.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.13.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.14.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.15.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.16.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.17.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.18.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.19.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.2.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.20.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.21.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.22.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.23.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.24.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.25.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.26.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.3.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.4.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.5.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.6.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.7.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.8.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.attn.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.attn.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.attn.qkv.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.attn.qkv.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.mlp.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.mlp.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.mlp.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.mlp.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.norm1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.norm1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.norm2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.blocks.9.norm2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.merger.linear_fc1.bias": "model-00004-of-00004.safetensors",
+    "model.visual.merger.linear_fc1.weight": "model-00004-of-00004.safetensors",
+    "model.visual.merger.linear_fc2.bias": "model-00004-of-00004.safetensors",
+    "model.visual.merger.linear_fc2.weight": "model-00004-of-00004.safetensors",
+    "model.visual.merger.norm.bias": "model-00004-of-00004.safetensors",
+    "model.visual.merger.norm.weight": "model-00004-of-00004.safetensors",
+    "model.visual.patch_embed.proj.bias": "model-00004-of-00004.safetensors",
+    "model.visual.patch_embed.proj.weight": "model-00004-of-00004.safetensors",
+    "model.visual.pos_embed.weight": "model-00004-of-00004.safetensors"
+  }
+}

a6_coupled_proposer_coder/round_4/preprocessor_config.json ADDED Viewed

	@@ -0,0 +1,21 @@

+{
+    "size": {
+        "longest_edge": 16777216,
+        "shortest_edge": 65536
+    },
+    "patch_size": 16,
+    "temporal_patch_size": 2,
+    "merge_size": 2,
+    "image_mean": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "image_std": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "processor_class": "Qwen3VLProcessor",
+    "image_processor_type": "Qwen2VLImageProcessorFast"
+}

a6_coupled_proposer_coder/round_4/processor_config.json ADDED Viewed

	@@ -0,0 +1,61 @@

+{
+  "image_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_processor_type": "Qwen2VLImageProcessor",
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "merge_size": 2,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "size": {
+      "longest_edge": 16777216,
+      "shortest_edge": 65536
+    },
+    "temporal_patch_size": 2
+  },
+  "processor_class": "Qwen3VLProcessor",
+  "video_processor": {
+    "do_convert_rgb": true,
+    "do_normalize": true,
+    "do_rescale": true,
+    "do_resize": true,
+    "do_sample_frames": true,
+    "fps": 2,
+    "image_mean": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "image_std": [
+      0.5,
+      0.5,
+      0.5
+    ],
+    "max_frames": 768,
+    "max_video_tokens": 768,
+    "merge_size": 2,
+    "min_frames": 4,
+    "patch_size": 16,
+    "resample": 3,
+    "rescale_factor": 0.00392156862745098,
+    "return_metadata": false,
+    "size": {
+      "longest_edge": 25165824,
+      "shortest_edge": 4096
+    },
+    "temporal_patch_size": 2,
+    "video_processor_type": "Qwen3VLVideoProcessor"
+  }
+}

a6_coupled_proposer_coder/round_4/tokenizer.json ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:06b9509352d2af50381ab2247e083b80d32d5c0aba91c272ca9ff729b6a0e523
+size 19989325

a6_coupled_proposer_coder/round_4/tokenizer_config.json ADDED Viewed

	@@ -0,0 +1,33 @@

+{
+  "add_prefix_space": false,
+  "audio_bos_token": "<|audio_start|>",
+  "audio_eos_token": "<|audio_end|>",
+  "audio_token": "<|audio_pad|>",
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "image_token": "<|image_pad|>",
+  "is_local": true,
+  "local_files_only": true,
+  "model_max_length": 262144,
+  "model_specific_special_tokens": {
+    "audio_bos_token": "<|audio_start|>",
+    "audio_eos_token": "<|audio_end|>",
+    "audio_token": "<|audio_pad|>",
+    "image_token": "<|image_pad|>",
+    "video_token": "<|video_pad|>",
+    "vision_bos_token": "<|vision_start|>",
+    "vision_eos_token": "<|vision_end|>"
+  },
+  "pad_token": "<|endoftext|>",
+  "pretokenize_regex": "(?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\\r\\n\\p{L}\\p{N}]?[\\p{L}\\p{M}]+|\\p{N}| ?[^\\s\\p{L}\\p{M}\\p{N}]+[\\r\\n]*|\\s*[\\r\\n]+|\\s+(?!\\S)|\\s+",
+  "processor_class": "Qwen3VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "video_token": "<|video_pad|>",
+  "vision_bos_token": "<|vision_start|>",
+  "vision_eos_token": "<|vision_end|>"
+}

a6_coupled_proposer_coder/round_4/trainer_state.json ADDED Viewed

	@@ -0,0 +1,2454 @@

+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 16.0,
+  "global_step": 160,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.2265625,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 1042.20703125,
+      "completions/min_length": 86.0,
+      "epoch": 0.00625,
+      "frac_reward_zero_std": 0.21875,
+      "grad_norm": 2.4948819750412157,
+      "kl": 0.0,
+      "learning_rate": 2e-07,
+      "loss": -7.450580596923828e-09,
+      "reward": 0.4296875,
+      "reward_std": 0.37282809615135193,
+      "rewards/MajorityAccuracy/mean": 0.4296875,
+      "rewards/MajorityAccuracy/std": 0.49600115418434143,
+      "step": 1,
+      "step_time": 163.8385738111101
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "epoch": 0.0125,
+      "grad_norm": 1.657529523161166,
+      "kl": 0.0,
+      "learning_rate": 4e-07,
+      "loss": -9.313225746154785e-10,
+      "step": 2,
+      "step_time": 20.133057564962655
+    },
+    {
+      "clip_ratio/high_max": 0.0007108147256076336,
+      "clip_ratio/high_mean": 0.00023320990294450894,
+      "clip_ratio/low_mean": 0.0003298914853075985,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0005631013846141286,
+      "epoch": 0.01875,
+      "grad_norm": 3.245535741827053,
+      "kl": 0.0002373412426095456,
+      "learning_rate": 6e-07,
+      "loss": 0.00040541961789131165,
+      "step": 3,
+      "step_time": 21.93793280282989
+    },
+    {
+      "clip_ratio/high_max": 0.0020238922152202576,
+      "clip_ratio/high_mean": 0.0005059730538050644,
+      "clip_ratio/low_mean": 1.52587890625e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0005212318465055432,
+      "epoch": 0.025,
+      "grad_norm": 8.45168368970901,
+      "kl": 0.00012406847599777393,
+      "learning_rate": 8e-07,
+      "loss": -0.00021903417655266821,
+      "step": 4,
+      "step_time": 26.180574710015208
+    },
+    {
+      "clip_ratio/high_max": 0.001447876333259046,
+      "clip_ratio/high_mean": 0.0004542936076177284,
+      "clip_ratio/low_mean": 0.00042316103645134717,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0008774546440690756,
+      "epoch": 0.03125,
+      "grad_norm": 2.3652065337298476,
+      "kl": 0.00021838758402736858,
+      "learning_rate": 1e-06,
+      "loss": 7.969420403242111e-05,
+      "step": 5,
+      "step_time": 26.502015486825258
+    },
+    {
+      "clip_ratio/high_max": 0.00114045110240113,
+      "clip_ratio/high_mean": 0.00033655310107860714,
+      "clip_ratio/low_mean": 0.00014544353689416312,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00048199663797277026,
+      "epoch": 0.0375,
+      "grad_norm": 2.0313050359600626,
+      "kl": 0.00021209513215580955,
+      "learning_rate": 9.998973021172564e-07,
+      "loss": -7.603707490488887e-05,
+      "step": 6,
+      "step_time": 26.756799736991525
+    },
+    {
+      "clip_ratio/high_max": 0.0024787351430859417,
+      "clip_ratio/high_mean": 0.0007585471030324697,
+      "clip_ratio/low_mean": 0.00037130792770767584,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0011298550307401456,
+      "epoch": 0.04375,
+      "grad_norm": 3.012363565384627,
+      "kl": 0.0002760400548140751,
+      "learning_rate": 9.99589250656446e-07,
+      "loss": 0.00024072491214610636,
+      "step": 7,
+      "step_time": 23.293301870580763
+    },
+    {
+      "clip_ratio/high_max": 0.0024818489473545924,
+      "clip_ratio/high_mean": 0.0008716796110093128,
+      "clip_ratio/low_mean": 0.0006540627582580782,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0015257423729053698,
+      "epoch": 0.05,
+      "grad_norm": 2.383036495637996,
+      "kl": 0.000303953847833327,
+      "learning_rate": 9.990759721625005e-07,
+      "loss": -0.00015883793821558356,
+      "step": 8,
+      "step_time": 26.441846468485892
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.21875,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 1030.55859375,
+      "completions/min_length": 64.0,
+      "epoch": 0.05625,
+      "frac_reward_zero_std": 0.09375,
+      "grad_norm": 2.650015797443978,
+      "kl": 0.0003404338931431994,
+      "learning_rate": 9.983576774858775e-07,
+      "loss": 3.38628888130188e-06,
+      "reward": 0.41796875,
+      "reward_std": 0.43175023794174194,
+      "rewards/MajorityAccuracy/mean": 0.41796875,
+      "rewards/MajorityAccuracy/std": 0.49419113993644714,
+      "step": 9,
+      "step_time": 147.1941068265587
+    },
+    {
+      "clip_ratio/high_max": 0.004852875717915595,
+      "clip_ratio/high_mean": 0.0017440347437513992,
+      "clip_ratio/low_mean": 0.00027581208632909693,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.002019846830080496,
+      "epoch": 0.0625,
+      "grad_norm": 3.852195321496234,
+      "kl": 0.0009214362144120969,
+      "learning_rate": 9.974346616959475e-07,
+      "loss": 8.394243195652962e-06,
+      "step": 10,
+      "step_time": 24.89420717069879
+    },
+    {
+      "clip_ratio/high_max": 0.006689656001981348,
+      "clip_ratio/high_mean": 0.001672414000495337,
+      "clip_ratio/low_mean": 0.0016366419495170703,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.003309056024590973,
+      "epoch": 0.06875,
+      "grad_norm": 3.3680565855371003,
+      "kl": 0.000946158317674417,
+      "learning_rate": 9.963073039597796e-07,
+      "loss": -0.0005626333877444267,
+      "step": 11,
+      "step_time": 22.159679376054555
+    },
+    {
+      "clip_ratio/high_max": 0.0017580909698153846,
+      "clip_ratio/high_mean": 0.0007649015151400818,
+      "clip_ratio/low_mean": 0.000452398920970154,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0012173004433861934,
+      "epoch": 0.075,
+      "grad_norm": 2.1617021048661655,
+      "kl": 0.0008646452333778143,
+      "learning_rate": 9.949760673863846e-07,
+      "loss": 0.0003258834476582706,
+      "step": 12,
+      "step_time": 26.71327840304002
+    },
+    {
+      "clip_ratio/high_max": 0.004952458228217438,
+      "clip_ratio/high_mean": 0.0016019873801269569,
+      "clip_ratio/low_mean": 0.001232328220794443,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0028343156300252303,
+      "epoch": 0.08125,
+      "grad_norm": 2.447706578172051,
+      "kl": 0.0008392176532652229,
+      "learning_rate": 9.934414988364722e-07,
+      "loss": 0.00033177435398101807,
+      "step": 13,
+      "step_time": 25.948615881148726
+    },
+    {
+      "clip_ratio/high_max": 0.0029115989163983613,
+      "clip_ratio/high_mean": 0.0009901563316816464,
+      "clip_ratio/low_mean": 0.0006090636597946286,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0015992200060281903,
+      "epoch": 0.0875,
+      "grad_norm": 2.008173208288417,
+      "kl": 0.0007079485076246783,
+      "learning_rate": 9.917042286978063e-07,
+      "loss": 0.0003741949039977044,
+      "step": 14,
+      "step_time": 23.360323519445956
+    },
+    {
+      "clip_ratio/high_max": 0.012555334600619972,
+      "clip_ratio/high_mean": 0.00330374130862765,
+      "clip_ratio/low_mean": 0.002639048485434614,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0059427898086141795,
+      "epoch": 0.09375,
+      "grad_norm": 3.305676874618124,
+      "kl": 0.00142505721305497,
+      "learning_rate": 9.897649706262473e-07,
+      "loss": 0.000563240610063076,
+      "step": 15,
+      "step_time": 25.672556331846863
+    },
+    {
+      "clip_ratio/high_max": 0.0066364804515615106,
+      "clip_ratio/high_mean": 0.002373140727286227,
+      "clip_ratio/low_mean": 0.002086479653371498,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00445962039520964,
+      "epoch": 0.1,
+      "grad_norm": 2.2309755631234944,
+      "kl": 0.0009486579729127698,
+      "learning_rate": 9.87624521252587e-07,
+      "loss": 0.0005562715232372284,
+      "step": 16,
+      "step_time": 25.876083539333194
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.2421875,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 1070.6171875,
+      "completions/min_length": 78.0,
+      "epoch": 0.10625,
+      "frac_reward_zero_std": 0.21875,
+      "grad_norm": 2.1220000250595956,
+      "kl": 0.0011913466223631985,
+      "learning_rate": 9.852837598553008e-07,
+      "loss": 1.1898265256604645e-05,
+      "reward": 0.49609375,
+      "reward_std": 0.34785041213035583,
+      "rewards/MajorityAccuracy/mean": 0.49609375,
+      "rewards/MajorityAccuracy/std": 0.5009641051292419,
+      "step": 17,
+      "step_time": 147.8995105009526
+    },
+    {
+      "clip_ratio/high_max": 0.0036880426050629467,
+      "clip_ratio/high_mean": 0.0013163789190002717,
+      "clip_ratio/low_mean": 0.0007237006793729961,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.002040079634753056,
+      "epoch": 0.1125,
+      "grad_norm": 2.612455244777466,
+      "kl": 0.0013241683845990337,
+      "learning_rate": 9.827436479993468e-07,
+      "loss": 7.487181574106216e-05,
+      "step": 18,
+      "step_time": 23.288427517283708
+    },
+    {
+      "clip_ratio/high_max": 0.0026426693075336516,
+      "clip_ratio/high_mean": 0.0007245297310873866,
+      "clip_ratio/low_mean": 0.00038544931339856703,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0011099790499429218,
+      "epoch": 0.11875,
+      "grad_norm": 1.6672980208950583,
+      "kl": 0.001954952036612667,
+      "learning_rate": 9.800052291411629e-07,
+      "loss": 0.00014653940161224455,
+      "step": 19,
+      "step_time": 25.393264647573233
+    },
+    {
+      "clip_ratio/high_max": 0.0013053904112894088,
+      "clip_ratio/high_mean": 0.0003980219116783701,
+      "clip_ratio/low_mean": 0.00045412078907247633,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.000852142708026804,
+      "epoch": 0.125,
+      "grad_norm": 1.4750769427134491,
+      "kl": 0.002176202440750785,
+      "learning_rate": 9.770696282000244e-07,
+      "loss": -0.0002451106265652925,
+      "step": 20,
+      "step_time": 26.852842475287616
+    },
+    {
+      "clip_ratio/high_max": 0.0053911065551801585,
+      "clip_ratio/high_mean": 0.002353618319830275,
+      "clip_ratio/low_mean": 0.0008100247941911221,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.003163643108564429,
+      "epoch": 0.13125,
+      "grad_norm": 2.114707423608449,
+      "kl": 0.0023128017492126673,
+      "learning_rate": 9.739380510959364e-07,
+      "loss": 6.768200546503067e-05,
+      "step": 21,
+      "step_time": 23.195452659856528
+    },
+    {
+      "clip_ratio/high_max": 0.007696756656514481,
+      "clip_ratio/high_mean": 0.0021988705193507485,
+      "clip_ratio/low_mean": 0.0006327308219624683,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0028316013485891744,
+      "epoch": 0.1375,
+      "grad_norm": 2.1833975255203084,
+      "kl": 0.005357940564863384,
+      "learning_rate": 9.706117842542516e-07,
+      "loss": 0.0003631244180724025,
+      "step": 22,
+      "step_time": 23.440711108502
+    },
+    {
+      "clip_ratio/high_max": 0.006723283266182989,
+      "clip_ratio/high_mean": 0.0024870144698070362,
+      "clip_ratio/low_mean": 0.0014291088227764703,
+      "clip_ratio/low_min": 0.00018310546875,
+      "clip_ratio/region_mean": 0.003916123241651803,
+      "epoch": 0.14375,
+      "grad_norm": 1.4889206403008606,
+      "kl": 0.0025270933692809194,
+      "learning_rate": 9.670921940772186e-07,
+      "loss": 0.0004541119560599327,
+      "step": 23,
+      "step_time": 26.78288444550708
+    },
+    {
+      "clip_ratio/high_max": 0.008217017370043322,
+      "clip_ratio/high_mean": 0.003103725779510569,
+      "clip_ratio/low_mean": 0.003669688376248814,
+      "clip_ratio/low_min": 0.0007122346723917872,
+      "clip_ratio/region_mean": 0.006773414177587256,
+      "epoch": 0.15,
+      "grad_norm": 2.9938921851557736,
+      "kl": 0.00434474807116203,
+      "learning_rate": 9.633807263826744e-07,
+      "loss": 0.0010416172444820404,
+      "step": 24,
+      "step_time": 20.712629717309028
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.16015625,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 928.25390625,
+      "completions/min_length": 50.0,
+      "epoch": 0.15625,
+      "frac_reward_zero_std": 0.25,
+      "grad_norm": 6.588847710005894,
+      "kl": 0.0037565334350802004,
+      "learning_rate": 9.594789058101153e-07,
+      "loss": 3.7568988773273304e-05,
+      "reward": 0.45703125,
+      "reward_std": 0.3595416843891144,
+      "rewards/MajorityAccuracy/mean": 0.45703125,
+      "rewards/MajorityAccuracy/std": 0.4991260766983032,
+      "step": 25,
+      "step_time": 145.2621974251233
+    },
+    {
+      "clip_ratio/high_max": 0.000614987569861114,
+      "clip_ratio/high_mean": 0.0001537468924652785,
+      "clip_ratio/low_mean": 0.0002970371806441108,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0004507840731093893,
+      "epoch": 0.1625,
+      "grad_norm": 2.1848976676518737,
+      "kl": 0.0026939279487123713,
+      "learning_rate": 9.55388335194388e-07,
+      "loss": 9.332012268714607e-05,
+      "step": 26,
+      "step_time": 26.39336795685813
+    },
+    {
+      "clip_ratio/high_max": 0.0041775892896112055,
+      "clip_ratio/high_mean": 0.001461159401515033,
+      "clip_ratio/low_mean": 0.0005215978799242293,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001982757283258252,
+      "epoch": 0.16875,
+      "grad_norm": 3.8088696025870945,
+      "kl": 0.004527848825091496,
+      "learning_rate": 9.511106949072587e-07,
+      "loss": 0.0008781757205724716,
+      "step": 27,
+      "step_time": 22.164075868204236
+    },
+    {
+      "clip_ratio/high_max": 0.003748827235540375,
+      "clip_ratio/high_mean": 0.001282011107832659,
+      "clip_ratio/low_mean": 0.00121434242464602,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0024963535252027214,
+      "epoch": 0.175,
+      "grad_norm": 2.2176417361972107,
+      "kl": 0.0032980641699396074,
+      "learning_rate": 9.466477421671295e-07,
+      "loss": -0.00023835033061914146,
+      "step": 28,
+      "step_time": 21.932904582470655
+    },
+    {
+      "clip_ratio/high_max": 0.00926662604615558,
+      "clip_ratio/high_mean": 0.002815361487591872,
+      "clip_ratio/low_mean": 0.0009079608644242398,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0037233223629300483,
+      "epoch": 0.18125,
+      "grad_norm": 4.815319875189683,
+      "kl": 0.005299718410242349,
+      "learning_rate": 9.420013103171891e-07,
+      "loss": 0.0004975396441295743,
+      "step": 29,
+      "step_time": 22.089062666986138
+    },
+    {
+      "clip_ratio/high_max": 0.0070428808248834684,
+      "clip_ratio/high_mean": 0.0023250718259077985,
+      "clip_ratio/low_mean": 0.002221432419901248,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00454650420579128,
+      "epoch": 0.1875,
+      "grad_norm": 2.6399581469574067,
+      "kl": 0.005113313905894756,
+      "learning_rate": 9.37173308072291e-07,
+      "loss": 0.00018920726142823696,
+      "step": 30,
+      "step_time": 25.573548825457692
+    },
+    {
+      "clip_ratio/high_max": 0.0019238487293478101,
+      "clip_ratio/high_mean": 0.00048096218233695254,
+      "clip_ratio/low_mean": 0.0009596132294973359,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0014405754336621612,
+      "epoch": 0.19375,
+      "grad_norm": 1.2033094152059522,
+      "kl": 0.007598980329930782,
+      "learning_rate": 9.321657187348688e-07,
+      "loss": 0.00026289757806807756,
+      "step": 31,
+      "step_time": 24.864173350390047
+    },
+    {
+      "clip_ratio/high_max": 0.007623435943969525,
+      "clip_ratio/high_mean": 0.0021445742786454502,
+      "clip_ratio/low_mean": 0.00227949908003211,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.004424073325935751,
+      "epoch": 0.2,
+      "grad_norm": 2.1222683788892374,
+      "kl": 0.004097250406630337,
+      "learning_rate": 9.269805993802128e-07,
+      "loss": 0.00014629280485678464,
+      "step": 32,
+      "step_time": 26.54124335804954
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0703125,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 753.765625,
+      "completions/min_length": 76.0,
+      "epoch": 0.20625,
+      "frac_reward_zero_std": 0.15625,
+      "grad_norm": 2.1443559349300916,
+      "kl": 0.0038687701453454792,
+      "learning_rate": 9.216200800114411e-07,
+      "loss": 3.870110958814621e-05,
+      "reward": 0.46875,
+      "reward_std": 0.3813767433166504,
+      "rewards/MajorityAccuracy/mean": 0.46875,
+      "rewards/MajorityAccuracy/std": 0.5,
+      "step": 33,
+      "step_time": 135.032489140518
+    },
+    {
+      "clip_ratio/high_max": 0.0015807089512236416,
+      "clip_ratio/high_mean": 0.0005623164979624562,
+      "clip_ratio/low_mean": 9.483773101237603e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0006571542289748322,
+      "epoch": 0.2125,
+      "grad_norm": 1.6911422311196171,
+      "kl": 0.0041776195575948805,
+      "learning_rate": 9.160863626845119e-07,
+      "loss": -2.2527317923959345e-05,
+      "step": 34,
+      "step_time": 22.61290714284405
+    },
+    {
+      "clip_ratio/high_max": 0.005630045692669228,
+      "clip_ratio/high_mean": 0.001810444075090345,
+      "clip_ratio/low_mean": 0.0009982540723285638,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.002808698161970824,
+      "epoch": 0.21875,
+      "grad_norm": 3.439194146059957,
+      "kl": 0.007427943288348615,
+      "learning_rate": 9.103817206036382e-07,
+      "loss": 0.00041345134377479553,
+      "step": 35,
+      "step_time": 24.33624576544389
+    },
+    {
+      "clip_ratio/high_max": 0.0031738707766635343,
+      "clip_ratio/high_mean": 0.0009971047802537214,
+      "clip_ratio/low_mean": 0.0029018928325967863,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0038989976019365713,
+      "epoch": 0.225,
+      "grad_norm": 3.8947932705879293,
+      "kl": 0.008607811469119042,
+      "learning_rate": 9.045084971874737e-07,
+      "loss": 0.00044621527194976807,
+      "step": 36,
+      "step_time": 19.540111381560564
+    },
+    {
+      "clip_ratio/high_max": 0.006717185664456338,
+      "clip_ratio/high_mean": 0.0022196566642378457,
+      "clip_ratio/low_mean": 0.0015517963765887544,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0037714530481025577,
+      "epoch": 0.23125,
+      "grad_norm": 3.7202267484145453,
+      "kl": 0.011996103217825294,
+      "learning_rate": 8.984691051064574e-07,
+      "loss": 0.00036742910742759705,
+      "step": 37,
+      "step_time": 20.89849931327626
+    },
+    {
+      "clip_ratio/high_max": 0.0032334416755475104,
+      "clip_ratio/high_mean": 0.0010627256997395307,
+      "clip_ratio/low_mean": 0.0009181908171740361,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001980916509637609,
+      "epoch": 0.2375,
+      "grad_norm": 1.8463583482359773,
+      "kl": 0.005899310810491443,
+      "learning_rate": 8.922660252917087e-07,
+      "loss": 0.00025007486692629755,
+      "step": 38,
+      "step_time": 21.086580173578113
+    },
+    {
+      "clip_ratio/high_max": 0.006920372572494671,
+      "clip_ratio/high_mean": 0.002037580678006634,
+      "clip_ratio/low_mean": 0.0017151577994809486,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.003752738528419286,
+      "epoch": 0.24375,
+      "grad_norm": 1.7552680382822088,
+      "kl": 0.005844740488100797,
+      "learning_rate": 8.859018059158809e-07,
+      "loss": 0.0003241325612179935,
+      "step": 39,
+      "step_time": 24.87123668147251
+    },
+    {
+      "clip_ratio/high_max": 0.012263727170648053,
+      "clip_ratio/high_mean": 0.004224648408126086,
+      "clip_ratio/low_mean": 0.002508509933250025,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0067331582540646195,
+      "epoch": 0.25,
+      "grad_norm": 9.304384106535478,
+      "kl": 0.005357540008844808,
+      "learning_rate": 8.793790613463954e-07,
+      "loss": 0.00031978264451026917,
+      "step": 40,
+      "step_time": 21.271742640063167
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.26171875,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 1080.0078125,
+      "completions/min_length": 71.0,
+      "epoch": 0.25625,
+      "frac_reward_zero_std": 0.21875,
+      "grad_norm": 0.8949994099979419,
+      "kl": 0.0039113592356443405,
+      "learning_rate": 8.727004710714798e-07,
+      "loss": 3.9123719034250826e-05,
+      "reward": 0.36328125,
+      "reward_std": 0.36717280745506287,
+      "rewards/MajorityAccuracy/mean": 0.36328125,
+      "rewards/MajorityAccuracy/std": 0.48188698291778564,
+      "step": 41,
+      "step_time": 149.57459840737283
+    },
+    {
+      "clip_ratio/high_max": 0.0018131314864149317,
+      "clip_ratio/high_mean": 0.0005602864912361838,
+      "clip_ratio/low_mean": 0.0007346291968133301,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0012949156516697258,
+      "epoch": 0.2625,
+      "grad_norm": 2.937995201643706,
+      "kl": 0.005919241346418858,
+      "learning_rate": 8.658687785994578e-07,
+      "loss": 0.0001759100705385208,
+      "step": 42,
+      "step_time": 24.733856969513
+    },
+    {
+      "clip_ratio/high_max": 0.004005019698524848,
+      "clip_ratio/high_mean": 0.0013717183537664823,
+      "clip_ratio/low_mean": 0.0004137563519179821,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0017854747165984008,
+      "epoch": 0.26875,
+      "grad_norm": 3.358411488798062,
+      "kl": 0.00971053633838892,
+      "learning_rate": 8.588867903317394e-07,
+      "loss": 0.0004914960009045899,
+      "step": 43,
+      "step_time": 25.25034177256748
+    },
+    {
+      "clip_ratio/high_max": 0.0023918378865346313,
+      "clip_ratio/high_mean": 0.0009083567274501547,
+      "clip_ratio/low_mean": 4.952456583851017e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009578812932886649,
+      "epoch": 0.275,
+      "grad_norm": 2.154391963699547,
+      "kl": 0.009355776710435748,
+      "learning_rate": 8.51757374409974e-07,
+      "loss": 0.00033119524596259,
+      "step": 44,
+      "step_time": 26.79395919991657
+    },
+    {
+      "clip_ratio/high_max": 0.008318877080455422,
+      "clip_ratio/high_mean": 0.002747912658378482,
+      "clip_ratio/low_mean": 0.002709913213038817,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.005457825900521129,
+      "epoch": 0.28125,
+      "grad_norm": 3.9007640059587074,
+      "kl": 0.016479505342431366,
+      "learning_rate": 8.444834595378433e-07,
+      "loss": 0.00011155754327774048,
+      "step": 45,
+      "step_time": 20.400427412241697
+    },
+    {
+      "clip_ratio/high_max": 0.0033263834193348885,
+      "clip_ratio/high_mean": 0.0012927508214488626,
+      "clip_ratio/low_mean": 0.0007943358577904291,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0020870867156190798,
+      "epoch": 0.2875,
+      "grad_norm": 2.1665922511254836,
+      "kl": 0.004096979973837733,
+      "learning_rate": 8.370680337779736e-07,
+      "loss": 0.00014656968414783478,
+      "step": 46,
+      "step_time": 25.663893251214176
+    },
+    {
+      "clip_ratio/high_max": 0.009889261171338148,
+      "clip_ratio/high_mean": 0.0029782287201669533,
+      "clip_ratio/low_mean": 0.002653518953593448,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.005631747597362846,
+      "epoch": 0.29375,
+      "grad_norm": 2.994879145207738,
+      "kl": 0.008689777634572238,
+      "learning_rate": 8.295141433244659e-07,
+      "loss": -2.554524689912796e-05,
+      "step": 47,
+      "step_time": 25.610705588012934
+    },
+    {
+      "clip_ratio/high_max": 0.003830469591775909,
+      "clip_ratio/high_mean": 0.001211669405165594,
+      "clip_ratio/low_mean": 0.002041514453594573,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.003253183822380379,
+      "epoch": 0.3,
+      "grad_norm": 1.4967057207974048,
+      "kl": 0.006160397257190198,
+      "learning_rate": 8.218248912515442e-07,
+      "loss": 0.0004630539915524423,
+      "step": 48,
+      "step_time": 26.082106488756835
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.19140625,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 853.484375,
+      "completions/min_length": 116.0,
+      "epoch": 0.30625,
+      "frac_reward_zero_std": 0.21875,
+      "grad_norm": 2.84571384381275,
+      "kl": 0.006253765372093767,
+      "learning_rate": 8.140034362388397e-07,
+      "loss": 6.249541183933616e-05,
+      "reward": 0.4453125,
+      "reward_std": 0.3779529631137848,
+      "rewards/MajorityAccuracy/mean": 0.4453125,
+      "rewards/MajorityAccuracy/std": 0.49797382950782776,
+      "step": 49,
+      "step_time": 148.27969375718385
+    },
+    {
+      "clip_ratio/high_max": 0.0016141473752213642,
+      "clip_ratio/high_mean": 0.00046877692875568755,
+      "clip_ratio/low_mean": 0.0004432078349054791,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009119847709371243,
+      "epoch": 0.3125,
+      "grad_norm": 1.8949041389570025,
+      "kl": 0.008384999528061599,
+      "learning_rate": 8.060529912738314e-07,
+      "loss": 0.00029877485940232873,
+      "step": 50,
+      "step_time": 26.757319296244532
+    },
+    {
+      "clip_ratio/high_max": 0.002731624583248049,
+      "clip_ratio/high_mean": 0.0010619606473483145,
+      "clip_ratio/low_mean": 0.0021565191709669307,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0032184797892114148,
+      "epoch": 0.31875,
+      "grad_norm": 4.886249457282203,
+      "kl": 0.014953569683711976,
+      "learning_rate": 7.979768223319785e-07,
+      "loss": 0.0009001195430755615,
+      "step": 51,
+      "step_time": 18.958329448010772
+    },
+    {
+      "clip_ratio/high_max": 0.011782487330492586,
+      "clip_ratio/high_mean": 0.004424701459356584,
+      "clip_ratio/low_mean": 0.0010226197337033227,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.005447321163956076,
+      "epoch": 0.325,
+      "grad_norm": 4.375530529674499,
+      "kl": 0.011569544789381325,
+      "learning_rate": 7.897782470350849e-07,
+      "loss": 0.0007416053558699787,
+      "step": 52,
+      "step_time": 27.01575354859233
+    },
+    {
+      "clip_ratio/high_max": 0.0035406009264988825,
+      "clip_ratio/high_mean": 0.0008851502316247206,
+      "clip_ratio/low_mean": 0.00045313400551094674,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0013382842116698157,
+      "epoch": 0.33125,
+      "grad_norm": 1.7422491335337535,
+      "kl": 0.008388124697376043,
+      "learning_rate": 7.814606332884488e-07,
+      "loss": 0.000389587483368814,
+      "step": 53,
+      "step_time": 20.50143823120743
+    },
+    {
+      "clip_ratio/high_max": 0.007253997813677415,
+      "clip_ratio/high_mean": 0.0020655274638556875,
+      "clip_ratio/low_mean": 0.0023398070261464454,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.004405334504554048,
+      "epoch": 0.3375,
+      "grad_norm": 2.711626682578941,
+      "kl": 0.008482773322612047,
+      "learning_rate": 7.730273978973552e-07,
+      "loss": 0.00012007914483547211,
+      "step": 54,
+      "step_time": 25.3449998954311
+    },
+    {
+      "clip_ratio/high_max": 0.01576588026364334,
+      "clip_ratio/high_mean": 0.004776546840730589,
+      "clip_ratio/low_mean": 0.0014334710940602235,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.006210017862031236,
+      "epoch": 0.34375,
+      "grad_norm": 2.4222887354779536,
+      "kl": 0.008644929097499698,
+      "learning_rate": 7.644820051634812e-07,
+      "loss": 0.00028908593230880797,
+      "step": 55,
+      "step_time": 26.37621946260333
+    },
+    {
+      "clip_ratio/high_max": 0.005447258648928255,
+      "clip_ratio/high_mean": 0.002151813212549314,
+      "clip_ratio/low_mean": 0.0016162003739736974,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0037680136447306722,
+      "epoch": 0.35,
+      "grad_norm": 2.425695533101298,
+      "kl": 0.007398323039524257,
+      "learning_rate": 7.558279654617912e-07,
+      "loss": -0.00010295771062374115,
+      "step": 56,
+      "step_time": 24.880898064933717
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.25,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 1138.70703125,
+      "completions/min_length": 116.0,
+      "epoch": 0.35625,
+      "frac_reward_zero_std": 0.21875,
+      "grad_norm": 4.345355989287623,
+      "kl": 0.006837411900050938,
+      "learning_rate": 7.470688337985029e-07,
+      "loss": 6.843451410531998e-05,
+      "reward": 0.4140625,
+      "reward_std": 0.363634318113327,
+      "rewards/MajorityAccuracy/mean": 0.4140625,
+      "rewards/MajorityAccuracy/std": 0.4935242235660553,
+      "step": 57,
+      "step_time": 145.63945607095957
+    },
+    {
+      "clip_ratio/high_max": 0.0009372660861117765,
+      "clip_ratio/high_mean": 0.00025116288088611327,
+      "clip_ratio/low_mean": 0.00022465386064141057,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00047581674152752385,
+      "epoch": 0.3625,
+      "grad_norm": 1.2771265936011102,
+      "kl": 0.008176219998858869,
+      "learning_rate": 7.382082083507225e-07,
+      "loss": 8.943965076468885e-05,
+      "step": 58,
+      "step_time": 27.00241111172363
+    },
+    {
+      "clip_ratio/high_max": 0.002357497149205301,
+      "clip_ratio/high_mean": 0.0008055432754190406,
+      "clip_ratio/low_mean": 0.00013465497249853797,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009401982551935362,
+      "epoch": 0.36875,
+      "grad_norm": 1.765846565927869,
+      "kl": 0.008605382754467428,
+      "learning_rate": 7.292497289883432e-07,
+      "loss": 4.587249713949859e-05,
+      "step": 59,
+      "step_time": 26.87382673751563
+    },
+    {
+      "clip_ratio/high_max": 0.0016107427945826203,
+      "clip_ratio/high_mean": 0.00040268569864565507,
+      "clip_ratio/low_mean": 0.0008237541042035446,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001226439795573242,
+      "epoch": 0.375,
+      "grad_norm": 1.864679956048009,
+      "kl": 0.005315476737450808,
+      "learning_rate": 7.201970757788171e-07,
+      "loss": 7.54315551603213e-05,
+      "step": 60,
+      "step_time": 27.178127312101424
+    },
+    {
+      "clip_ratio/high_max": 0.0033362207032041624,
+      "clip_ratio/high_mean": 0.0011864773332490586,
+      "clip_ratio/low_mean": 0.0010575770938885398,
+      "clip_ratio/low_min": 6.480041338363662e-05,
+      "clip_ratio/region_mean": 0.002244054412585683,
+      "epoch": 0.38125,
+      "grad_norm": 1.7748726022356631,
+      "kl": 0.007739013643004,
+      "learning_rate": 7.110539674754159e-07,
+      "loss": 0.00042327120900154114,
+      "step": 61,
+      "step_time": 26.630333130713552
+    },
+    {
+      "clip_ratio/high_max": 0.002767086654785089,
+      "clip_ratio/high_mean": 0.0009690922124718782,
+      "clip_ratio/low_mean": 0.0004379973943287041,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001407089632266434,
+      "epoch": 0.3875,
+      "grad_norm": 2.328437987217767,
+      "kl": 0.006793023436330259,
+      "learning_rate": 7.018241599895973e-07,
+      "loss": 0.00041560083627700806,
+      "step": 62,
+      "step_time": 23.025342358276248
+    },
+    {
+      "clip_ratio/high_max": 0.0064293480245396495,
+      "clip_ratio/high_mean": 0.0021303338144207373,
+      "clip_ratio/low_mean": 0.002147082064766437,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00427741592284292,
+      "epoch": 0.39375,
+      "grad_norm": 4.189829453015119,
+      "kl": 0.01166293234564364,
+      "learning_rate": 6.925114448481088e-07,
+      "loss": 0.0003764871507883072,
+      "step": 63,
+      "step_time": 23.54398933891207
+    },
+    {
+      "clip_ratio/high_max": 0.002945535903563723,
+      "clip_ratio/high_mean": 0.0009959359231288545,
+      "clip_ratio/low_mean": 0.0012102770997444168,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.002206212986493483,
+      "epoch": 0.4,
+      "grad_norm": 1.388841033864293,
+      "kl": 0.012384369503706694,
+      "learning_rate": 6.831196476354614e-07,
+      "loss": 3.0071889341343194e-05,
+      "step": 64,
+      "step_time": 24.78142129909247
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.234375,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 1049.4453125,
+      "completions/min_length": 73.0,
+      "epoch": 0.40625,
+      "frac_reward_zero_std": 0.125,
+      "grad_norm": 1.9615099517610284,
+      "kl": 0.009762316360138357,
+      "learning_rate": 6.7365262642241e-07,
+      "loss": 9.765475988388062e-05,
+      "reward": 0.42578125,
+      "reward_std": 0.38600224256515503,
+      "rewards/MajorityAccuracy/mean": 0.42578125,
+      "rewards/MajorityAccuracy/std": 0.49542948603630066,
+      "step": 65,
+      "step_time": 153.1635134355165
+    },
+    {
+      "clip_ratio/high_max": 0.0021649636037182063,
+      "clip_ratio/high_mean": 0.0005765915775555186,
+      "clip_ratio/low_mean": 0.0006460675795096904,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0012226591643411666,
+      "epoch": 0.4125,
+      "grad_norm": 2.436500850735864,
+      "kl": 0.009108558122534305,
+      "learning_rate": 6.641142701810931e-07,
+      "loss": 0.00023490394232794642,
+      "step": 66,
+      "step_time": 25.7647008812055
+    },
+    {
+      "clip_ratio/high_max": 0.0019843479894916527,
+      "clip_ratio/high_mean": 0.0005958885867585195,
+      "clip_ratio/low_mean": 0.0016195859425351955,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0022154745456646197,
+      "epoch": 0.41875,
+      "grad_norm": 2.2165468819759253,
+      "kl": 0.0129739010008052,
+      "learning_rate": 6.545084971874736e-07,
+      "loss": 7.150229066610336e-05,
+      "step": 67,
+      "step_time": 25.18778609763831
+    },
+    {
+      "clip_ratio/high_max": 0.0012610003323061392,
+      "clip_ratio/high_mean": 0.0005804878419439774,
+      "clip_ratio/low_mean": 0.0007361232856055722,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0013166111348255072,
+      "epoch": 0.425,
+      "grad_norm": 2.6142611789811894,
+      "kl": 0.01180325448513031,
+      "learning_rate": 6.448392534117461e-07,
+      "loss": 0.00016710534691810608,
+      "step": 68,
+      "step_time": 24.997970247175545
+    },
+    {
+      "clip_ratio/high_max": 0.004578250314807519,
+      "clip_ratio/high_mean": 0.0015133345368667506,
+      "clip_ratio/low_mean": 0.00022913943212188315,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0017424739562557079,
+      "epoch": 0.43125,
+      "grad_norm": 2.55162101650617,
+      "kl": 0.010152662289328873,
+      "learning_rate": 6.351105108973644e-07,
+      "loss": 0.0004765090998262167,
+      "step": 69,
+      "step_time": 26.724271754268557
+    },
+    {
+      "clip_ratio/high_max": 0.006386901892255992,
+      "clip_ratio/high_mean": 0.002437861039652489,
+      "clip_ratio/low_mean": 0.0010362229550082702,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.003474084020126611,
+      "epoch": 0.4375,
+      "grad_norm": 2.7997072232390687,
+      "kl": 0.008059154963120818,
+      "learning_rate": 6.253262661293602e-07,
+      "loss": 4.028715193271637e-05,
+      "step": 70,
+      "step_time": 23.456438838969916
+    },
+    {
+      "clip_ratio/high_max": 0.007344981422647834,
+      "clip_ratio/high_mean": 0.002575244099716656,
+      "clip_ratio/low_mean": 0.0003820424899458885,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.002957286560558714,
+      "epoch": 0.44375,
+      "grad_norm": 2.056839918267484,
+      "kl": 0.016195324482396245,
+      "learning_rate": 6.154905383926216e-07,
+      "loss": 2.898721140809357e-05,
+      "step": 71,
+      "step_time": 26.24803021363914
+    },
+    {
+      "clip_ratio/high_max": 0.008093027456197888,
+      "clip_ratio/high_mean": 0.0021458791015902534,
+      "clip_ratio/low_mean": 0.0012922896603413392,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.003438168714637868,
+      "epoch": 0.45,
+      "grad_norm": 3.2063051096264052,
+      "kl": 0.009228141803760082,
+      "learning_rate": 6.056073681208037e-07,
+      "loss": -2.620436134748161e-05,
+      "step": 72,
+      "step_time": 26.251589470077306
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.203125,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 947.39453125,
+      "completions/min_length": 87.0,
+      "epoch": 0.45625,
+      "frac_reward_zero_std": 0.1875,
+      "grad_norm": 4.311601508467849,
+      "kl": 0.018673236714676023,
+      "learning_rate": 5.956808152365532e-07,
+      "loss": 0.00018669944256544113,
+      "reward": 0.50390625,
+      "reward_std": 0.35890138149261475,
+      "rewards/MajorityAccuracy/mean": 0.50390625,
+      "rewards/MajorityAccuracy/std": 0.5009641051292419,
+      "step": 73,
+      "step_time": 137.2573599969037
+    },
+    {
+      "clip_ratio/high_max": 0.0020141963759670034,
+      "clip_ratio/high_mean": 0.0005733225298172329,
+      "clip_ratio/low_mean": 0.0005751417011197191,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001148464230936952,
+      "epoch": 0.4625,
+      "grad_norm": 2.8589028874994455,
+      "kl": 0.012905186624266207,
+      "learning_rate": 5.857149574837268e-07,
+      "loss": 0.000270012766122818,
+      "step": 74,
+      "step_time": 20.146482759155333
+    },
+    {
+      "clip_ratio/high_max": 0.00146819790097652,
+      "clip_ratio/high_mean": 0.00048211969806288835,
+      "clip_ratio/low_mean": 0.0004252123508194927,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009073320507013705,
+      "epoch": 0.46875,
+      "grad_norm": 1.9455743451217764,
+      "kl": 0.009977033245377243,
+      "learning_rate": 5.757138887522883e-07,
+      "loss": -9.786341252038255e-05,
+      "step": 75,
+      "step_time": 26.58821861539036
+    },
+    {
+      "clip_ratio/high_max": 0.0026698422443587333,
+      "clip_ratio/high_mean": 0.0010161099453398492,
+      "clip_ratio/low_mean": 0.00011929161155421752,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0011354015587130561,
+      "epoch": 0.475,
+      "grad_norm": 1.5327858354251216,
+      "kl": 0.009577100514434278,
+      "learning_rate": 5.656817173965732e-07,
+      "loss": -7.47969897929579e-06,
+      "step": 76,
+      "step_time": 24.50140861282125
+    },
+    {
+      "clip_ratio/high_max": 0.008411933085881174,
+      "clip_ratio/high_mean": 0.00263846293091774,
+      "clip_ratio/low_mean": 0.0014296364824986085,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.004068099427968264,
+      "epoch": 0.48125,
+      "grad_norm": 2.9797708994267027,
+      "kl": 0.012293737847357988,
+      "learning_rate": 5.556225645476118e-07,
+      "loss": -1.2236414477229118e-05,
+      "step": 77,
+      "step_time": 24.62603202648461
+    },
+    {
+      "clip_ratio/high_max": 0.006029650190612301,
+      "clip_ratio/high_mean": 0.0016494994270033203,
+      "clip_ratio/low_mean": 0.0011041818943340331,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.002753681328613311,
+      "epoch": 0.4875,
+      "grad_norm": 1.7090556894513085,
+      "kl": 0.009889445966109633,
+      "learning_rate": 5.455405624202032e-07,
+      "loss": 0.00037296852678991854,
+      "step": 78,
+      "step_time": 24.75735648209229
+    },
+    {
+      "clip_ratio/high_max": 0.010418544872663915,
+      "clip_ratio/high_mean": 0.0038979911187198013,
+      "clip_ratio/low_mean": 0.0018484789034118876,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.005746469949372113,
+      "epoch": 0.49375,
+      "grad_norm": 4.037382544363377,
+      "kl": 0.01967799966223538,
+      "learning_rate": 5.354398526154365e-07,
+      "loss": 0.0014543216675519943,
+      "step": 79,
+      "step_time": 26.585793574806303
+    },
+    {
+      "clip_ratio/high_max": 0.005426783551229164,
+      "clip_ratio/high_mean": 0.0017168765916721895,
+      "clip_ratio/low_mean": 0.0019863902998622507,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.003703266876982525,
+      "epoch": 0.5,
+      "grad_norm": 3.388032480205817,
+      "kl": 0.010190736036747694,
+      "learning_rate": 5.253245844193564e-07,
+      "loss": 0.0002476843656040728,
+      "step": 80,
+      "step_time": 26.116106947883964
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.15234375,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 809.828125,
+      "completions/min_length": 72.0,
+      "epoch": 0.50625,
+      "frac_reward_zero_std": 0.1875,
+      "grad_norm": 1.9692062464900126,
+      "kl": 0.010196742718107998,
+      "learning_rate": 5.151989130984714e-07,
+      "loss": 0.00010196678340435028,
+      "reward": 0.515625,
+      "reward_std": 0.37492266297340393,
+      "rewards/MajorityAccuracy/mean": 0.515625,
+      "rewards/MajorityAccuracy/std": 0.5007347464561462,
+      "step": 81,
+      "step_time": 145.29902549274266
+    },
+    {
+      "clip_ratio/high_max": 0.0011776122846640646,
+      "clip_ratio/high_mean": 0.00029440307116601616,
+      "clip_ratio/low_mean": 9.477532876189798e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00038917840720387176,
+      "epoch": 0.5125,
+      "grad_norm": 1.7967826671047165,
+      "kl": 0.012629186850972474,
+      "learning_rate": 5.050669981928055e-07,
+      "loss": 0.00023962398699950427,
+      "step": 82,
+      "step_time": 26.43158908886835
+    },
+    {
+      "clip_ratio/high_max": 0.0038752385298721492,
+      "clip_ratio/high_mean": 0.0010286755277775228,
+      "clip_ratio/low_mean": 0.0008494138783134986,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0018780893988150638,
+      "epoch": 0.51875,
+      "grad_norm": 3.219558599603571,
+      "kl": 0.02482287958264351,
+      "learning_rate": 4.949330018071946e-07,
+      "loss": 0.00037262329715304077,
+      "step": 83,
+      "step_time": 18.13717239582911
+    },
+    {
+      "clip_ratio/high_max": 0.002925579232396558,
+      "clip_ratio/high_mean": 0.0010481392600922845,
+      "clip_ratio/low_mean": 0.001861810857008095,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.002909950133471284,
+      "epoch": 0.525,
+      "grad_norm": 2.662564509976143,
+      "kl": 0.015946133295074105,
+      "learning_rate": 4.848010869015287e-07,
+      "loss": -0.00023903569672256708,
+      "step": 84,
+      "step_time": 25.880740280728787
+    },
+    {
+      "clip_ratio/high_max": 0.0030080022988840938,
+      "clip_ratio/high_mean": 0.0008800100622465834,
+      "clip_ratio/low_mean": 0.001163839151558932,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0020438492065295577,
+      "epoch": 0.53125,
+      "grad_norm": 1.8252086049061167,
+      "kl": 0.012977503007277846,
+      "learning_rate": 4.746754155806437e-07,
+      "loss": 0.0001932010054588318,
+      "step": 85,
+      "step_time": 23.641466598026454
+    },
+    {
+      "clip_ratio/high_max": 0.005081150462501682,
+      "clip_ratio/high_mean": 0.001432089560694294,
+      "clip_ratio/low_mean": 0.0001886522713903105,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0016207418302656151,
+      "epoch": 0.5375,
+      "grad_norm": 1.279620919652818,
+      "kl": 0.012630559038370848,
+      "learning_rate": 4.645601473845635e-07,
+      "loss": 7.726644980721176e-05,
+      "step": 86,
+      "step_time": 23.286530560348183
+    },
+    {
+      "clip_ratio/high_max": 0.010533664171816781,
+      "clip_ratio/high_mean": 0.003236073018342722,
+      "clip_ratio/low_mean": 0.0014654831757070497,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.004701556186773814,
+      "epoch": 0.54375,
+      "grad_norm": 5.586528421743571,
+      "kl": 0.028507261537015438,
+      "learning_rate": 4.544594375797968e-07,
+      "loss": 0.00033564493060112,
+      "step": 87,
+      "step_time": 25.636804479174316
+    },
+    {
+      "clip_ratio/high_max": 0.00558452436234802,
+      "clip_ratio/high_mean": 0.0024914592795539647,
+      "clip_ratio/low_mean": 0.0012656931612582412,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0037571524444501847,
+      "epoch": 0.55,
+      "grad_norm": 3.4263698173506127,
+      "kl": 0.016050332342274487,
+      "learning_rate": 4.443774354523882e-07,
+      "loss": 0.0005893371999263763,
+      "step": 88,
+      "step_time": 22.41457739379257
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.16796875,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 982.0546875,
+      "completions/min_length": 107.0,
+      "epoch": 0.55625,
+      "frac_reward_zero_std": 0.125,
+      "grad_norm": 3.0112890111928636,
+      "kl": 0.01688663789536804,
+      "learning_rate": 4.3431828260342675e-07,
+      "loss": 0.00016889255493879318,
+      "reward": 0.42578125,
+      "reward_std": 0.40373584628105164,
+      "rewards/MajorityAccuracy/mean": 0.42578125,
+      "rewards/MajorityAccuracy/std": 0.49542948603630066,
+      "step": 89,
+      "step_time": 144.54777716472745
+    },
+    {
+      "clip_ratio/high_max": 0.0030155514978105202,
+      "clip_ratio/high_mean": 0.0008354696583410259,
+      "clip_ratio/low_mean": 0.0006010088909533806,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0014364785165525973,
+      "epoch": 0.5625,
+      "grad_norm": 3.1562515252098184,
+      "kl": 0.014181873761117458,
+      "learning_rate": 4.242861112477118e-07,
+      "loss": 0.0003285938873887062,
+      "step": 90,
+      "step_time": 23.794549088459462
+    },
+    {
+      "clip_ratio/high_max": 0.0023466227648896165,
+      "clip_ratio/high_mean": 0.0006657508365606191,
+      "clip_ratio/low_mean": 0.00044393487769411877,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001109685719711706,
+      "epoch": 0.56875,
+      "grad_norm": 2.27765856026404,
+      "kl": 0.013853005017153919,
+      "learning_rate": 4.1428504251627325e-07,
+      "loss": 6.390126509359106e-05,
+      "step": 91,
+      "step_time": 25.794307083357126
+    },
+    {
+      "clip_ratio/high_max": 0.0059207171070738696,
+      "clip_ratio/high_mean": 0.0017383649301336845,
+      "clip_ratio/low_mean": 0.0007327396888285875,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.002471104613505304,
+      "epoch": 0.575,
+      "grad_norm": 3.508937365829386,
+      "kl": 0.01564521132968366,
+      "learning_rate": 4.0431918476344685e-07,
+      "loss": -6.9753936259076e-05,
+      "step": 92,
+      "step_time": 25.83836164837703
+    },
+    {
+      "clip_ratio/high_max": 0.0043889999215025455,
+      "clip_ratio/high_mean": 0.0015458536217920482,
+      "clip_ratio/low_mean": 0.0013975928122818004,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0029434464231599122,
+      "epoch": 0.58125,
+      "grad_norm": 3.2339774321941444,
+      "kl": 0.019556229817681015,
+      "learning_rate": 3.943926318791963e-07,
+      "loss": 0.00023604463785886765,
+      "step": 93,
+      "step_time": 20.796353679616004
+    },
+    {
+      "clip_ratio/high_max": 0.0031784588063601404,
+      "clip_ratio/high_mean": 0.0010276267566950992,
+      "clip_ratio/low_mean": 0.0008507244801876368,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0018783512496156618,
+      "epoch": 0.5875,
+      "grad_norm": 2.7768263613498427,
+      "kl": 0.01943243178538978,
+      "learning_rate": 3.845094616073783e-07,
+      "loss": 0.00039792805910110474,
+      "step": 94,
+      "step_time": 26.096686519216746
+    },
+    {
+      "clip_ratio/high_max": 0.0020874959664070047,
+      "clip_ratio/high_mean": 0.000797484217400779,
+      "clip_ratio/low_mean": 0.0015258951498253737,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.002323379318113439,
+      "epoch": 0.59375,
+      "grad_norm": 4.970709604981883,
+      "kl": 0.0190582568757236,
+      "learning_rate": 3.7467373387063964e-07,
+      "loss": 0.00018784776329994202,
+      "step": 95,
+      "step_time": 22.22055195691064
+    },
+    {
+      "clip_ratio/high_max": 0.0020099272951483727,
+      "clip_ratio/high_mean": 0.0007283311861101538,
+      "clip_ratio/low_mean": 0.00047070125583559275,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001199032434669789,
+      "epoch": 0.6,
+      "grad_norm": 1.769485689867788,
+      "kl": 0.010319368331693113,
+      "learning_rate": 3.648894891026358e-07,
+      "loss": 0.00022432717378251255,
+      "step": 96,
+      "step_time": 26.905255460646003
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.31640625,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 1170.7734375,
+      "completions/min_length": 80.0,
+      "epoch": 0.60625,
+      "frac_reward_zero_std": 0.28125,
+      "grad_norm": 2.4175795644181175,
+      "kl": 0.0166850034147501,
+      "learning_rate": 3.5516074658825397e-07,
+      "loss": 0.00016679891268722713,
+      "reward": 0.3359375,
+      "reward_std": 0.3104974031448364,
+      "rewards/MajorityAccuracy/mean": 0.3359375,
+      "rewards/MajorityAccuracy/std": 0.4732423424720764,
+      "step": 97,
+      "step_time": 153.81195211596787
+    },
+    {
+      "clip_ratio/high_max": 0.00025188917061313987,
+      "clip_ratio/high_mean": 6.297229265328497e-05,
+      "clip_ratio/low_mean": 0.00025032674057001714,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0003132990332233021,
+      "epoch": 0.6125,
+      "grad_norm": 3.981478659973993,
+      "kl": 0.016192107344977558,
+      "learning_rate": 3.454915028125263e-07,
+      "loss": 0.00011750031262636185,
+      "step": 98,
+      "step_time": 24.058187837246805
+    },
+    {
+      "clip_ratio/high_max": 0.0041076012421399355,
+      "clip_ratio/high_mean": 0.0011370459396857768,
+      "clip_ratio/low_mean": 0.0005712539968953934,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0017082999474951066,
+      "epoch": 0.61875,
+      "grad_norm": 3.3606441551813844,
+      "kl": 0.02088077669031918,
+      "learning_rate": 3.3588572981890684e-07,
+      "loss": 0.0003605122328735888,
+      "step": 99,
+      "step_time": 26.74686147272587
+    },
+    {
+      "clip_ratio/high_max": 0.003076607157709077,
+      "clip_ratio/high_mean": 0.0009877026895992458,
+      "clip_ratio/low_mean": 0.0014022702089278027,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0023899728839751333,
+      "epoch": 0.625,
+      "grad_norm": 3.703836582155767,
+      "kl": 0.020691176410764456,
+      "learning_rate": 3.263473735775899e-07,
+      "loss": 0.00048251455882564187,
+      "step": 100,
+      "step_time": 24.410050339065492
+    },
+    {
+      "clip_ratio/high_max": 0.0007768440991640091,
+      "clip_ratio/high_mean": 0.00019421102479100227,
+      "clip_ratio/low_mean": 0.0004475738824112341,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0006417849072022364,
+      "epoch": 0.63125,
+      "grad_norm": 1.0405324596485002,
+      "kl": 0.01027990982402116,
+      "learning_rate": 3.1688035236453865e-07,
+      "loss": 9.03074542293325e-05,
+      "step": 101,
+      "step_time": 27.012250141706318
+    },
+    {
+      "clip_ratio/high_max": 0.001766232802765444,
+      "clip_ratio/high_mean": 0.0004826766235055402,
+      "clip_ratio/low_mean": 0.000812798421975458,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001295475049118977,
+      "epoch": 0.6375,
+      "grad_norm": 2.3629690356030193,
+      "kl": 0.013320311205461621,
+      "learning_rate": 3.0748855515189096e-07,
+      "loss": 0.00011905861902050674,
+      "step": 102,
+      "step_time": 26.793108568992466
+    },
+    {
+      "clip_ratio/high_max": 0.004927876914734952,
+      "clip_ratio/high_mean": 0.0013098994131723884,
+      "clip_ratio/low_mean": 0.0006434042225009762,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0019533036029315554,
+      "epoch": 0.64375,
+      "grad_norm": 3.1453714371325234,
+      "kl": 0.023051016498357058,
+      "learning_rate": 2.981758400104028e-07,
+      "loss": -0.0001538160431664437,
+      "step": 103,
+      "step_time": 25.60162579547614
+    },
+    {
+      "clip_ratio/high_max": 0.0013140094160917215,
+      "clip_ratio/high_mean": 0.00040682316284801345,
+      "clip_ratio/low_mean": 0.0003354709333507344,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0007422940907417797,
+      "epoch": 0.65,
+      "grad_norm": 1.4563989721457766,
+      "kl": 0.013541532098315656,
+      "learning_rate": 2.8894603252458403e-07,
+      "loss": 0.00021299337095115334,
+      "step": 104,
+      "step_time": 26.841613350901753
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.109375,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 779.234375,
+      "completions/min_length": 98.0,
+      "epoch": 0.65625,
+      "frac_reward_zero_std": 0.21875,
+      "grad_norm": 3.018574940260952,
+      "kl": 0.012074064929038286,
+      "learning_rate": 2.7980292422118277e-07,
+      "loss": 0.00012070871889591217,
+      "reward": 0.5234375,
+      "reward_std": 0.3456100821495056,
+      "rewards/MajorityAccuracy/mean": 0.5234375,
+      "rewards/MajorityAccuracy/std": 0.5004287362098694,
+      "step": 105,
+      "step_time": 137.950611715205
+    },
+    {
+      "clip_ratio/high_max": 0.0008582974405726418,
+      "clip_ratio/high_mean": 0.00028223791741766036,
+      "clip_ratio/low_mean": 0.0006597023966605775,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009419403068022802,
+      "epoch": 0.6625,
+      "grad_norm": 2.5323484086271986,
+      "kl": 0.01592361438088119,
+      "learning_rate": 2.70750271011657e-07,
+      "loss": 0.00014694873243570328,
+      "step": 106,
+      "step_time": 20.913985884282738
+    },
+    {
+      "clip_ratio/high_max": 0.0031772764632478356,
+      "clip_ratio/high_mean": 0.0012124349377700128,
+      "clip_ratio/low_mean": 0.0006998565149842761,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001912291430926416,
+      "epoch": 0.66875,
+      "grad_norm": 3.8208769947648054,
+      "kl": 0.021139377495273948,
+      "learning_rate": 2.6179179164927754e-07,
+      "loss": 0.0003456994891166687,
+      "step": 107,
+      "step_time": 19.98063875362277
+    },
+    {
+      "clip_ratio/high_max": 0.007625806916621514,
+      "clip_ratio/high_mean": 0.0023037118262436707,
+      "clip_ratio/low_mean": 0.00023529671307187527,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0025390085320395883,
+      "epoch": 0.675,
+      "grad_norm": 2.7459932980917245,
+      "kl": 0.02396414300892502,
+      "learning_rate": 2.529311662014972e-07,
+      "loss": 0.0004357607685960829,
+      "step": 108,
+      "step_time": 19.291070685256273
+    },
+    {
+      "clip_ratio/high_max": 0.0006314874190138653,
+      "clip_ratio/high_mean": 0.00015787185475346632,
+      "clip_ratio/low_mean": 0.0007376416360784788,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0008955134908319451,
+      "epoch": 0.68125,
+      "grad_norm": 3.9894385827882752,
+      "kl": 0.02056608721613884,
+      "learning_rate": 2.441720345382089e-07,
+      "loss": 0.00015445894678123295,
+      "step": 109,
+      "step_time": 26.904767704196274
+    },
+    {
+      "clip_ratio/high_max": 0.0023343147477135062,
+      "clip_ratio/high_mean": 0.0008092573771136813,
+      "clip_ratio/low_mean": 0.0001792291659512557,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009884865467029158,
+      "epoch": 0.6875,
+      "grad_norm": 1.7398301179174354,
+      "kl": 0.015284773078747094,
+      "learning_rate": 2.355179948365189e-07,
+      "loss": 0.000241497706156224,
+      "step": 110,
+      "step_time": 23.822653929702938
+    },
+    {
+      "clip_ratio/high_max": 0.0043646538542816415,
+      "clip_ratio/high_mean": 0.0013040807280049194,
+      "clip_ratio/low_mean": 0.00042293619117117487,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001727016904624179,
+      "epoch": 0.69375,
+      "grad_norm": 2.0780520719348177,
+      "kl": 0.02477374137379229,
+      "learning_rate": 2.26972602102645e-07,
+      "loss": 0.0004963691462762654,
+      "step": 111,
+      "step_time": 19.866626985371113
+    },
+    {
+      "clip_ratio/high_max": 0.003003162652021274,
+      "clip_ratio/high_mean": 0.0008253730047727004,
+      "clip_ratio/low_mean": 0.0006029279211361427,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0014283009331848007,
+      "epoch": 0.7,
+      "grad_norm": 3.24478308294139,
+      "kl": 0.028604315826669335,
+      "learning_rate": 2.1853936671155127e-07,
+      "loss": 0.0003909283550456166,
+      "step": 112,
+      "step_time": 21.591148674488068
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.16015625,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 879.7421875,
+      "completions/min_length": 74.0,
+      "epoch": 0.70625,
+      "frac_reward_zero_std": 0.15625,
+      "grad_norm": 6.658070020732353,
+      "kl": 0.036980316042900085,
+      "learning_rate": 2.1022175296491512e-07,
+      "loss": 0.00036988314241170883,
+      "reward": 0.46484375,
+      "reward_std": 0.39598357677459717,
+      "rewards/MajorityAccuracy/mean": 0.46484375,
+      "rewards/MajorityAccuracy/std": 0.49973952770233154,
+      "step": 113,
+      "step_time": 142.96141126286238
+    },
+    {
+      "clip_ratio/high_max": 0.0014637597487308085,
+      "clip_ratio/high_mean": 0.000481263043184299,
+      "clip_ratio/low_mean": 0.00041441356734139845,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0008956766105256975,
+      "epoch": 0.7125,
+      "grad_norm": 8.239820161150428,
+      "kl": 0.019435668364167213,
+      "learning_rate": 2.0202317766802152e-07,
+      "loss": 0.00045881234109401703,
+      "step": 114,
+      "step_time": 25.18920602602884
+    },
+    {
+      "clip_ratio/high_max": 0.0023478658331441693,
+      "clip_ratio/high_mean": 0.0005869664582860423,
+      "clip_ratio/low_mean": 0.00038224041782086715,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009692068779258989,
+      "epoch": 0.71875,
+      "grad_norm": 2.1851513876409285,
+      "kl": 0.019137667259201407,
+      "learning_rate": 1.9394700872616853e-07,
+      "loss": 0.00043937709415331483,
+      "step": 115,
+      "step_time": 25.560870122164488
+    },
+    {
+      "clip_ratio/high_max": 0.0034241390239913017,
+      "clip_ratio/high_mean": 0.0012948774456162937,
+      "clip_ratio/low_mean": 0.0004274611837900011,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0017223386275873054,
+      "epoch": 0.725,
+      "grad_norm": 3.5401846203849647,
+      "kl": 0.018627459881827235,
+      "learning_rate": 1.8599656376116024e-07,
+      "loss": 0.0005011362954974174,
+      "step": 116,
+      "step_time": 25.937369257211685
+    },
+    {
+      "clip_ratio/high_max": 0.004354332166258246,
+      "clip_ratio/high_mean": 0.0013940030548837967,
+      "clip_ratio/low_mean": 0.0003801272760028951,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0017741303490765858,
+      "epoch": 0.73125,
+      "grad_norm": 3.154644252050376,
+      "kl": 0.019527108990587294,
+      "learning_rate": 1.7817510874845582e-07,
+      "loss": 0.0002474130014888942,
+      "step": 117,
+      "step_time": 23.91040014475584
+    },
+    {
+      "clip_ratio/high_max": 0.001685697672655806,
+      "clip_ratio/high_mean": 0.0005439734450192191,
+      "clip_ratio/low_mean": 0.0006831052378402092,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0012270786901353858,
+      "epoch": 0.7375,
+      "grad_norm": 3.6530097535653168,
+      "kl": 0.029583486495539546,
+      "learning_rate": 1.7048585667553412e-07,
+      "loss": 0.0005061350530013442,
+      "step": 118,
+      "step_time": 17.492355132941157
+    },
+    {
+      "clip_ratio/high_max": 0.00501804355008062,
+      "clip_ratio/high_mean": 0.0014844201650703326,
+      "clip_ratio/low_mean": 0.00026047303254017606,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0017448931866965722,
+      "epoch": 0.74375,
+      "grad_norm": 1.8704430385044934,
+      "kl": 0.01812749938108027,
+      "learning_rate": 1.6293196622202632e-07,
+      "loss": 0.00025342879234813154,
+      "step": 119,
+      "step_time": 24.89409090578556
+    },
+    {
+      "clip_ratio/high_max": 0.0035653329541673884,
+      "clip_ratio/high_mean": 0.0012953894802194554,
+      "clip_ratio/low_mean": 0.000925862324947957,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0022212517651496455,
+      "epoch": 0.75,
+      "grad_norm": 2.8846194966759593,
+      "kl": 0.01596821763087064,
+      "learning_rate": 1.555165404621567e-07,
+      "loss": 1.0266900062561035e-05,
+      "step": 120,
+      "step_time": 26.672801355365664
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.0859375,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 726.515625,
+      "completions/min_length": 65.0,
+      "epoch": 0.75625,
+      "frac_reward_zero_std": 0.25,
+      "grad_norm": 3.138569444319796,
+      "kl": 0.021689745597541332,
+      "learning_rate": 1.4824262559002592e-07,
+      "loss": 0.00021702982485294342,
+      "reward": 0.5703125,
+      "reward_std": 0.35771310329437256,
+      "rewards/MajorityAccuracy/mean": 0.5703125,
+      "rewards/MajorityAccuracy/std": 0.49600115418434143,
+      "step": 121,
+      "step_time": 137.4545444007963
+    },
+    {
+      "clip_ratio/high_max": 0.0023843290691729635,
+      "clip_ratio/high_mean": 0.0007209256073110737,
+      "clip_ratio/low_mean": 3.0517578125e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0007514431854360737,
+      "epoch": 0.7625,
+      "grad_norm": 1.717259440583081,
+      "kl": 0.018575544003397226,
+      "learning_rate": 1.4111320966826057e-07,
+      "loss": 3.1667048460803926e-05,
+      "step": 122,
+      "step_time": 26.88156923558563
+    },
+    {
+      "clip_ratio/high_max": 0.00517063592269551,
+      "clip_ratio/high_mean": 0.001851247259764932,
+      "clip_ratio/low_mean": 0.0013705154706258327,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.003221762686735019,
+      "epoch": 0.76875,
+      "grad_norm": 5.245107622760394,
+      "kl": 0.017161862226203084,
+      "learning_rate": 1.3413122140054217e-07,
+      "loss": 0.00026083365082740784,
+      "step": 123,
+      "step_time": 24.468251530546695
+    },
+    {
+      "clip_ratio/high_max": 0.003291189219453372,
+      "clip_ratio/high_mean": 0.0011967202626692597,
+      "clip_ratio/low_mean": 0.00022166025883052498,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0014183805214997847,
+      "epoch": 0.775,
+      "grad_norm": 4.657908948330544,
+      "kl": 0.029344928450882435,
+      "learning_rate": 1.272995289285202e-07,
+      "loss": 0.00028617773205041885,
+      "step": 124,
+      "step_time": 17.811280873138458
+    },
+    {
+      "clip_ratio/high_max": 0.0022341277945088223,
+      "clip_ratio/high_mean": 0.0009065001795534045,
+      "clip_ratio/low_mean": 4.57763671875e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009522765467409045,
+      "epoch": 0.78125,
+      "grad_norm": 2.054761323933606,
+      "kl": 0.017973641399294138,
+      "learning_rate": 1.2062093865360457e-07,
+      "loss": 0.0002617223362904042,
+      "step": 125,
+      "step_time": 24.6447251169011
+    },
+    {
+      "clip_ratio/high_max": 0.0020182943262625486,
+      "clip_ratio/high_mean": 0.0005727306051994674,
+      "clip_ratio/low_mean": 3.265412669861689e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0006053847246221267,
+      "epoch": 0.7875,
+      "grad_norm": 3.201003351257642,
+      "kl": 0.019170701852999628,
+      "learning_rate": 1.1409819408411897e-07,
+      "loss": 0.0003697032807394862,
+      "step": 126,
+      "step_time": 21.272151058539748
+    },
+    {
+      "clip_ratio/high_max": 0.0017024658154696226,
+      "clip_ratio/high_mean": 0.0004808284720638767,
+      "clip_ratio/low_mean": 0.0006267282951739617,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0011075567672378384,
+      "epoch": 0.79375,
+      "grad_norm": 2.7156462045995284,
+      "kl": 0.017814268707297742,
+      "learning_rate": 1.0773397470829143e-07,
+      "loss": 0.0002215724380221218,
+      "step": 127,
+      "step_time": 20.32942428952083
+    },
+    {
+      "clip_ratio/high_max": 0.0020331843552412465,
+      "clip_ratio/high_mean": 0.0005814811047457624,
+      "clip_ratio/low_mean": 0.000277302668109769,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0008587837655795738,
+      "epoch": 0.8,
+      "grad_norm": 2.042755764311668,
+      "kl": 0.021590395364910364,
+      "learning_rate": 1.0153089489354256e-07,
+      "loss": 0.0002392069873167202,
+      "step": 128,
+      "step_time": 22.698068849742413
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.12890625,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 871.69921875,
+      "completions/min_length": 102.0,
+      "epoch": 0.80625,
+      "frac_reward_zero_std": 0.09375,
+      "grad_norm": 2.440852574707602,
+      "kl": 0.016751970164477825,
+      "learning_rate": 9.549150281252632e-08,
+      "loss": 0.00016747601330280304,
+      "reward": 0.51953125,
+      "reward_std": 0.4266253709793091,
+      "rewards/MajorityAccuracy/mean": 0.51953125,
+      "rewards/MajorityAccuracy/std": 0.5005970597267151,
+      "step": 129,
+      "step_time": 144.12447233637795
+    },
+    {
+      "clip_ratio/high_max": 0.004734133632155135,
+      "clip_ratio/high_mean": 0.0016546842016396113,
+      "clip_ratio/low_mean": 0.00025372459640493616,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0019084087980445474,
+      "epoch": 0.8125,
+      "grad_norm": 3.0657824760367305,
+      "kl": 0.027150588808581233,
+      "learning_rate": 8.961827939636196e-08,
+      "loss": 0.0005136039108037949,
+      "step": 130,
+      "step_time": 25.771599777042866
+    },
+    {
+      "clip_ratio/high_max": 0.005116784042911604,
+      "clip_ratio/high_mean": 0.0018941605303552933,
+      "clip_ratio/low_mean": 0.0003850989232887514,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0022792594536440447,
+      "epoch": 0.81875,
+      "grad_norm": 5.0118147729311815,
+      "kl": 0.029101260006427765,
+      "learning_rate": 8.391363731548811e-08,
+      "loss": 0.0005198698490858078,
+      "step": 131,
+      "step_time": 22.27458281023428
+    },
+    {
+      "clip_ratio/high_max": 0.003801055281655863,
+      "clip_ratio/high_mean": 0.0011153990562888794,
+      "clip_ratio/low_mean": 0.0004712325844593579,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0015866316243773326,
+      "epoch": 0.825,
+      "grad_norm": 3.591702250318269,
+      "kl": 0.029601371381431818,
+      "learning_rate": 7.837991998855897e-08,
+      "loss": 0.00023897085338830948,
+      "step": 132,
+      "step_time": 19.367446350399405
+    },
+    {
+      "clip_ratio/high_max": 0.0034691827895585448,
+      "clip_ratio/high_mean": 0.0009231980729964562,
+      "clip_ratio/low_mean": 0.000660580575640779,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0015837786413612776,
+      "epoch": 0.83125,
+      "grad_norm": 4.544851133554289,
+      "kl": 0.02673310716636479,
+      "learning_rate": 7.301940061978722e-08,
+      "loss": 0.00024831295013427734,
+      "step": 133,
+      "step_time": 24.48156679002568
+    },
+    {
+      "clip_ratio/high_max": 0.0023423897073371336,
+      "clip_ratio/high_mean": 0.0006350258663587738,
+      "clip_ratio/low_mean": 0.0008150347275659442,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0014500605975626968,
+      "epoch": 0.8375,
+      "grad_norm": 2.1759468092777072,
+      "kl": 0.022145352326333523,
+      "learning_rate": 6.783428126513125e-08,
+      "loss": 0.0005504892906174064,
+      "step": 134,
+      "step_time": 26.15206282865256
+    },
+    {
+      "clip_ratio/high_max": 0.0016943305381573737,
+      "clip_ratio/high_mean": 0.0004235826345393434,
+      "clip_ratio/low_mean": 0.0005205086199566722,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009440912690479308,
+      "epoch": 0.84375,
+      "grad_norm": 2.951005986708551,
+      "kl": 0.023316672537475824,
+      "learning_rate": 6.282669192770895e-08,
+      "loss": 0.0004349341616034508,
+      "step": 135,
+      "step_time": 24.05402083043009
+    },
+    {
+      "clip_ratio/high_max": 0.000981546807452105,
+      "clip_ratio/high_mean": 0.00024538670186302625,
+      "clip_ratio/low_mean": 0.00027227395912632346,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0005176606609893497,
+      "epoch": 0.85,
+      "grad_norm": 1.7975766311606518,
+      "kl": 0.020484501728788018,
+      "learning_rate": 5.799868968281074e-08,
+      "loss": 0.00031424546614289284,
+      "step": 136,
+      "step_time": 26.2127573848702
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.15625,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 959.671875,
+      "completions/min_length": 108.0,
+      "epoch": 0.85625,
+      "frac_reward_zero_std": 0.125,
+      "grad_norm": 2.2532533664937078,
+      "kl": 0.012500592623837292,
+      "learning_rate": 5.335225783287051e-08,
+      "loss": 0.0001250015338882804,
+      "reward": 0.3515625,
+      "reward_std": 0.41174525022506714,
+      "rewards/MajorityAccuracy/mean": 0.3515625,
+      "rewards/MajorityAccuracy/std": 0.47839343547821045,
+      "step": 137,
+      "step_time": 147.4805017975159
+    },
+    {
+      "clip_ratio/high_max": 0.0022148576681502163,
+      "clip_ratio/high_mean": 0.0005970569945930038,
+      "clip_ratio/low_mean": 0.0008271495407825569,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.00142420653719455,
+      "epoch": 0.8625,
+      "grad_norm": 3.488520166897144,
+      "kl": 0.019230642123147845,
+      "learning_rate": 4.888930509274125e-08,
+      "loss": 0.00021617114543914795,
+      "step": 138,
+      "step_time": 25.614510369487107
+    },
+    {
+      "clip_ratio/high_max": 0.0009429741985513829,
+      "clip_ratio/high_mean": 0.00023574354963784572,
+      "clip_ratio/low_mean": 0.000815194653114304,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0010509381900192238,
+      "epoch": 0.86875,
+      "grad_norm": 4.027994593848877,
+      "kl": 0.02041008265223354,
+      "learning_rate": 4.4611664805611794e-08,
+      "loss": 0.00045841559767723083,
+      "step": 139,
+      "step_time": 26.15382046625018
+    },
+    {
+      "clip_ratio/high_max": 0.002020225321757607,
+      "clip_ratio/high_mean": 0.0006303428417595569,
+      "clip_ratio/low_mean": 0.0007968145910126623,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0014271574746089755,
+      "epoch": 0.875,
+      "grad_norm": 4.178894146921362,
+      "kl": 0.018593456130474806,
+      "learning_rate": 4.0521094189884696e-08,
+      "loss": 0.00018082372844219208,
+      "step": 140,
+      "step_time": 21.279825566336513
+    },
+    {
+      "clip_ratio/high_max": 0.0016255849041044712,
+      "clip_ratio/high_mean": 0.0005020155840611551,
+      "clip_ratio/low_mean": 0.0010304174284101464,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001532433001557365,
+      "epoch": 0.88125,
+      "grad_norm": 2.7398354792738484,
+      "kl": 0.020787404500879347,
+      "learning_rate": 3.6619273617325695e-08,
+      "loss": 0.0004912940785288811,
+      "step": 141,
+      "step_time": 26.0451822206378
+    },
+    {
+      "clip_ratio/high_max": 0.001454809942515567,
+      "clip_ratio/high_mean": 0.00037896127469139174,
+      "clip_ratio/low_mean": 0.0005568857213802403,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009358469978906214,
+      "epoch": 0.8875,
+      "grad_norm": 1.4325508585196565,
+      "kl": 0.01993631524965167,
+      "learning_rate": 3.2907805922781476e-08,
+      "loss": 0.0003698942018672824,
+      "step": 142,
+      "step_time": 24.90206878911704
+    },
+    {
+      "clip_ratio/high_max": 0.003690522731631063,
+      "clip_ratio/high_mean": 0.0010646604023349937,
+      "clip_ratio/low_mean": 0.0003834096060018055,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0014480700156127568,
+      "epoch": 0.89375,
+      "grad_norm": 2.5046370294009623,
+      "kl": 0.015651732217520475,
+      "learning_rate": 2.9388215745748345e-08,
+      "loss": 0.00016962851805146784,
+      "step": 143,
+      "step_time": 24.436501033604145
+    },
+    {
+      "clip_ratio/high_max": 0.001820292032789439,
+      "clip_ratio/high_mean": 0.00045507300819735974,
+      "clip_ratio/low_mean": 0.0005248260131338611,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0009798990286071785,
+      "epoch": 0.9,
+      "grad_norm": 3.3637136234941045,
+      "kl": 0.020501769264228642,
+      "learning_rate": 2.6061948904063658e-08,
+      "loss": 0.0008498597890138626,
+      "step": 144,
+      "step_time": 22.07402921654284
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.13671875,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 747.16796875,
+      "completions/min_length": 75.0,
+      "epoch": 0.90625,
+      "frac_reward_zero_std": 0.3125,
+      "grad_norm": 4.0299199215043355,
+      "kl": 0.029384346213191748,
+      "learning_rate": 2.293037179997559e-08,
+      "loss": 0.00029393151635304093,
+      "reward": 0.54296875,
+      "reward_std": 0.31865015625953674,
+      "rewards/MajorityAccuracy/mean": 0.54296875,
+      "rewards/MajorityAccuracy/std": 0.4991260766983032,
+      "step": 145,
+      "step_time": 137.50296487333253
+    },
+    {
+      "clip_ratio/high_max": 0.001042176125338301,
+      "clip_ratio/high_mean": 0.0003358851608936675,
+      "clip_ratio/low_mean": 0.0005378368296078406,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0008737219905015081,
+      "epoch": 0.9125,
+      "grad_norm": 2.424240240614742,
+      "kl": 0.013947930769063532,
+      "learning_rate": 1.9994770858837107e-08,
+      "loss": 0.0002567193005234003,
+      "step": 146,
+      "step_time": 27.009563351981342
+    },
+    {
+      "clip_ratio/high_max": 0.0012769910536007956,
+      "clip_ratio/high_mean": 0.0003452029086474795,
+      "clip_ratio/low_mean": 0.00015468308629351668,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0004998859949409962,
+      "epoch": 0.91875,
+      "grad_norm": 2.2398265968932876,
+      "kl": 0.018517188145779073,
+      "learning_rate": 1.725635200065323e-08,
+      "loss": 0.0001949746219906956,
+      "step": 147,
+      "step_time": 19.58202040847391
+    },
+    {
+      "clip_ratio/high_max": 0.004231438564602286,
+      "clip_ratio/high_mean": 0.001127553143305704,
+      "clip_ratio/low_mean": 0.0002138541476597311,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0013414072982413927,
+      "epoch": 0.925,
+      "grad_norm": 2.591648722335122,
+      "kl": 0.025506752775982022,
+      "learning_rate": 1.4716240144699187e-08,
+      "loss": 0.0006493502296507359,
+      "step": 148,
+      "step_time": 21.4913614182733
+    },
+    {
+      "clip_ratio/high_max": 0.001788982073776424,
+      "clip_ratio/high_mean": 0.000447245518444106,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.000447245518444106,
+      "epoch": 0.93125,
+      "grad_norm": 4.367903801821457,
+      "kl": 0.017127352650277317,
+      "learning_rate": 1.2375478747413015e-08,
+      "loss": -4.179574898444116e-05,
+      "step": 149,
+      "step_time": 26.79510253900662
+    },
+    {
+      "clip_ratio/high_max": 0.0007933238084660843,
+      "clip_ratio/high_mean": 0.00026087263904628344,
+      "clip_ratio/low_mean": 0.00045319711352931336,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0007140697525755968,
+      "epoch": 0.9375,
+      "grad_norm": 2.652170808937731,
+      "kl": 0.012409659451805055,
+      "learning_rate": 1.0235029373752757e-08,
+      "loss": 0.0004564579576253891,
+      "step": 150,
+      "step_time": 26.206412605941296
+    },
+    {
+      "clip_ratio/high_max": 0.005205861358263064,
+      "clip_ratio/high_mean": 0.0021185084642638685,
+      "clip_ratio/low_mean": 0.00011003520921804011,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0022285436716629192,
+      "epoch": 0.94375,
+      "grad_norm": 3.7305503391504233,
+      "kl": 0.03152730828151107,
+      "learning_rate": 8.295771302193721e-09,
+      "loss": 0.0008722454076632857,
+      "step": 151,
+      "step_time": 21.658398329280317
+    },
+    {
+      "clip_ratio/high_max": 0.0021089340443722904,
+      "clip_ratio/high_mean": 0.00082281646609772,
+      "clip_ratio/low_mean": 0.0002170208827010356,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.001039837341522798,
+      "epoch": 0.95,
+      "grad_norm": 3.153495076059342,
+      "kl": 0.024417794542387128,
+      "learning_rate": 6.558501163527963e-09,
+      "loss": 0.000571819138713181,
+      "step": 152,
+      "step_time": 23.66858826391399
+    },
+    {
+      "clip_ratio/high_max": 0.0,
+      "clip_ratio/high_mean": 0.0,
+      "clip_ratio/low_mean": 0.0,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0,
+      "completions/clipped_ratio": 0.11328125,
+      "completions/max_length": 2048.0,
+      "completions/mean_length": 778.84375,
+      "completions/min_length": 97.0,
+      "epoch": 0.95625,
+      "frac_reward_zero_std": 0.21875,
+      "grad_norm": 2.5218953168391827,
+      "kl": 0.02316930191591382,
+      "learning_rate": 5.023932613615445e-09,
+      "loss": 0.00023159943521022797,
+      "reward": 0.49609375,
+      "reward_std": 0.3535032272338867,
+      "rewards/MajorityAccuracy/mean": 0.49609375,
+      "rewards/MajorityAccuracy/std": 0.5009641051292419,
+      "step": 153,
+      "step_time": 134.46895914664492
+    },
+    {
+      "clip_ratio/high_max": 0.004741020806250162,
+      "clip_ratio/high_mean": 0.001432830609701341,
+      "clip_ratio/low_mean": 0.00043661659583449364,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0018694471946218982,
+      "epoch": 0.9625,
+      "grad_norm": 3.3249985130617583,
+      "kl": 0.025198656134307384,
+      "learning_rate": 3.6926960402202666e-09,
+      "loss": 0.0004309096257202327,
+      "step": 154,
+      "step_time": 17.973916246090084
+    },
+    {
+      "clip_ratio/high_max": 0.00299190406803973,
+      "clip_ratio/high_mean": 0.000827695410407614,
+      "clip_ratio/low_mean": 5.5839200285845436e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.000883534617969417,
+      "epoch": 0.96875,
+      "grad_norm": 2.506025451031248,
+      "kl": 0.020713817328214645,
+      "learning_rate": 2.5653383040524224e-09,
+      "loss": 0.000238760287174955,
+      "step": 155,
+      "step_time": 25.598073198925704
+    },
+    {
+      "clip_ratio/high_max": 0.00203138125652913,
+      "clip_ratio/high_mean": 0.0005597125382337254,
+      "clip_ratio/low_mean": 5.32989579369314e-05,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0006130115034466144,
+      "epoch": 0.975,
+      "grad_norm": 1.7382386523613003,
+      "kl": 0.015180138754658401,
+      "learning_rate": 1.642322514122385e-09,
+      "loss": 0.00020493635383900255,
+      "step": 156,
+      "step_time": 26.868492702022195
+    },
+    {
+      "clip_ratio/high_max": 0.005774953562649898,
+      "clip_ratio/high_mean": 0.001593618519109441,
+      "clip_ratio/low_mean": 0.0008937172351579648,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.002487335790647194,
+      "epoch": 0.98125,
+      "grad_norm": 5.075481369706701,
+      "kl": 0.016250215005129576,
+      "learning_rate": 9.240278374995636e-10,
+      "loss": 0.00046332215424627066,
+      "step": 157,
+      "step_time": 17.51747187366709
+    },
+    {
+      "clip_ratio/high_max": 0.0020445640038815327,
+      "clip_ratio/high_mean": 0.0005379650265240343,
+      "clip_ratio/low_mean": 0.0003915676825272385,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.000929532705413294,
+      "epoch": 0.9875,
+      "grad_norm": 7.667457830603062,
+      "kl": 0.0192726394161582,
+      "learning_rate": 4.1074934355384006e-10,
+      "loss": 0.00034405026235617697,
+      "step": 158,
+      "step_time": 25.70570404175669
+    },
+    {
+      "clip_ratio/high_max": 0.0018336960638407618,
+      "clip_ratio/high_mean": 0.0005678192173945718,
+      "clip_ratio/low_mean": 0.0001944133618962951,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.0007622325792908669,
+      "epoch": 0.99375,
+      "grad_norm": 2.1286454862822235,
+      "kl": 0.014885155251249671,
+      "learning_rate": 1.0269788274358537e-10,
+      "loss": 0.0001212155184475705,
+      "step": 159,
+      "step_time": 25.935465823858976
+    },
+    {
+      "clip_ratio/high_max": 0.0011227851355215535,
+      "clip_ratio/high_mean": 0.00028069628388038836,
+      "clip_ratio/low_mean": 0.00041657347173895687,
+      "clip_ratio/low_min": 0.0,
+      "clip_ratio/region_mean": 0.000697269759257324,
+      "epoch": 1.0,
+      "grad_norm": 3.4576714755656375,
+      "kl": 0.020864882739260793,
+      "learning_rate": 0.0,
+      "loss": 0.0003305608406662941,
+      "step": 160,
+      "step_time": 21.303407743107527
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 160,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 16,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 0.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}