| [2026-05-20 01:49:25,315] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:100066] baseline 0.000GB () |
| [2026-05-20 01:49:25,315] [INFO] [axolotl.cli.config.load_cfg:341] [PID:100066] config: |
| { |
| "activation_offloading": false, |
| "adapter": "qlora", |
| "attn_implementation": "flash_attention_2", |
| "axolotl_config_path": "/content/configs/qwen36_27b_qlora_48k_openhands_segments_v2.yaml", |
| "base_model": "Qwen/Qwen3.6-27B", |
| "base_model_config": "Qwen/Qwen3.6-27B", |
| "batch_size": 8, |
| "bf16": true, |
| "capabilities": { |
| "bf16": true, |
| "compute_capability": "sm_80", |
| "fp8": false, |
| "n_gpu": 1, |
| "n_node": 1, |
| "tf32": true |
| }, |
| "context_parallel_size": 1, |
| "cut_cross_entropy": true, |
| "dataloader_num_workers": 2, |
| "dataloader_prefetch_factor": 2, |
| "dataset_num_proc": 12, |
| "dataset_prepared_path": "/content/outputs/last_run_prepared_48k_openhands_segments_v2", |
| "datasets": [ |
| { |
| "message_property_mappings": { |
| "content": "content", |
| "role": "role" |
| }, |
| "path": "/content/data/segments.train.jsonl", |
| "trust_remote_code": false, |
| "type": "input_output" |
| } |
| ], |
| "ddp": false, |
| "device": "cuda:0", |
| "device_map": "auto", |
| "dion_rank_fraction": 1.0, |
| "dion_rank_multiple_of": 1, |
| "eaft_alpha": 1.0, |
| "eaft_k": 20, |
| "embeddings_skip_upcast": true, |
| "env_capabilities": { |
| "torch_version": "2.10.0" |
| }, |
| "eot_tokens": [ |
| "<|im_end|>" |
| ], |
| "eval_batch_size": 1, |
| "eval_causal_lm_metrics": [ |
| "sacrebleu", |
| "comet", |
| "ter", |
| "chrf" |
| ], |
| "eval_max_new_tokens": 128, |
| "eval_sample_packing": true, |
| "eval_steps": 0.5, |
| "eval_table_size": 0, |
| "evals_per_epoch": 1, |
| "experimental_skip_move_to_device": true, |
| "flash_attention": true, |
| "fp16": false, |
| "generate_samples": false, |
| "generation_do_sample": true, |
| "generation_max_new_tokens": 50, |
| "generation_prompt_ratio": 0.5, |
| "generation_temperature": 0.7, |
| "gradient_accumulation_steps": 8, |
| "gradient_checkpointing": true, |
| "gradient_checkpointing_kwargs": { |
| "use_reentrant": false |
| }, |
| "greater_is_better": false, |
| "include_tkps": true, |
| "is_multimodal": true, |
| "layer_offloading": false, |
| "learning_rate": 0.0002, |
| "lisa_layers_attribute": "model.layers", |
| "load_best_model_at_end": true, |
| "load_in_4bit": true, |
| "load_in_8bit": false, |
| "local_rank": 0, |
| "logging_steps": 1, |
| "lora_alpha": 128, |
| "lora_dropout": 0.0, |
| "lora_mlp_kernel": true, |
| "lora_model_dir": "/content/outputs/qwen36-27b-qlora-48k-openhands", |
| "lora_o_kernel": true, |
| "lora_qkv_kernel": true, |
| "lora_r": 64, |
| "lora_target_modules": [ |
| "q_proj", |
| "k_proj", |
| "v_proj", |
| "o_proj", |
| "gate_proj", |
| "up_proj", |
| "down_proj" |
| ], |
| "loraplus_lr_embedding": 1e-06, |
| "lr_scheduler": "cosine", |
| "mean_resizing_embeddings": false, |
| "merge_method": "memory_efficient", |
| "metric_for_best_model": "eval_loss", |
| "micro_batch_size": 1, |
| "model_config_type": "qwen3_5", |
| "model_config_type_text": "qwen3_5_text", |
| "num_epochs": 2.0, |
| "num_generation_samples": 3, |
| "optimizer": "paged_adamw_8bit", |
| "otel_metrics_host": "localhost", |
| "otel_metrics_port": 8000, |
| "output_dir": "/content/outputs/qwen36-27b-qlora-48k-openhands/merged", |
| "pad_to_sequence_len": false, |
| "plugins": [ |
| "axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin" |
| ], |
| "pretrain_multipack_attn": true, |
| "processor_config": "Qwen/Qwen3.6-27B", |
| "profiler_steps_start": 0, |
| "qlora_sharded_model_loading": false, |
| "quantize_moe_experts": false, |
| "ray_num_workers": 1, |
| "resources_per_worker": { |
| "GPU": 1 |
| }, |
| "sample_packing": true, |
| "sample_packing_bin_size": 200, |
| "sample_packing_group_size": 100000, |
| "save_only_model": false, |
| "save_safetensors": true, |
| "save_steps": 0.25, |
| "save_total_limit": 6, |
| "saves_per_epoch": 2, |
| "seed": 42, |
| "sequence_len": 49152, |
| "shuffle_before_merging_datasets": false, |
| "shuffle_merged_datasets": true, |
| "skip_prepare_dataset": false, |
| "special_tokens": { |
| "eos_token": "<|im_end|>", |
| "pad_token": "<|endoftext|>" |
| }, |
| "streaming_multipack_buffer_size": 10000, |
| "strict": false, |
| "tensor_parallel_size": 1, |
| "test_datasets": [ |
| { |
| "message_property_mappings": { |
| "content": "content", |
| "role": "role" |
| }, |
| "path": "/content/data/segments.val.jsonl", |
| "trust_remote_code": false, |
| "type": "input_output" |
| } |
| ], |
| "tf32": true, |
| "tiled_mlp": true, |
| "tiled_mlp_use_original_mlp": true, |
| "tokenizer_config": "Qwen/Qwen3.6-27B", |
| "tokenizer_save_jinja_files": true, |
| "torch_dtype": "torch.bfloat16", |
| "train_on_inputs": false, |
| "trl": { |
| "async_prefetch": false, |
| "log_completions": false, |
| "mask_truncated_completions": false, |
| "ref_model_mixup_alpha": 0.9, |
| "ref_model_sync_steps": 64, |
| "replay_buffer_size": 0, |
| "replay_recompute_logps": true, |
| "reroll_max_groups": 1, |
| "reroll_start_fraction": 1.0, |
| "reward_num_workers": 1, |
| "scale_rewards": true, |
| "skip_zero_advantage_batches": true, |
| "sync_ref_model": false, |
| "use_data_producer": false, |
| "use_vllm": false, |
| "vllm_lora_sync": false, |
| "vllm_server_host": "0.0.0.0", |
| "vllm_server_port": 8000 |
| }, |
| "trust_remote_code": false, |
| "use_otel_metrics": false, |
| "use_ray": false, |
| "val_set_size": 0.0, |
| "vllm": { |
| "device": "auto", |
| "dtype": "auto", |
| "gpu_memory_utilization": 0.9, |
| "host": "0.0.0.0", |
| "port": 8000 |
| }, |
| "warmup_ratio": 0.03, |
| "weight_decay": 0.0, |
| "world_size": 1 |
| } |
| [2026-05-20 01:49:25,321] [INFO] [axolotl.integrations.base.register:380] [PID:100066] Attempting to load plugin: axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin |
| [2026-05-20 01:49:25,321] [INFO] [axolotl.integrations.base.register:383] [PID:100066] Plugin loaded successfully: axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin |
| [2026-05-20 01:49:25,392] [WARNING] [axolotl.utils.schemas.validation.check_sample_packing_without_attention:190] [PID:100066] sample_packing without flash, sdp, xformers, sage, or flex attention does not handle cross sample decontamination. |
| [2026-05-20 01:49:25,392] [WARNING] [axolotl.utils.schemas.validation.hint_sample_packing_padding:235] [PID:100066] `pad_to_sequence_len: true` is recommended when using sample_packing |
| [2026-05-20 01:49:25,540] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:100066] baseline 0.000GB () |
| [2026-05-20 01:49:25,541] [INFO] [axolotl.cli.config.load_cfg:341] [PID:100066] config: |
| { |
| "activation_offloading": false, |
| "adapter": "qlora", |
| "attn_implementation": "flash_attention_2", |
| "axolotl_config_path": "/content/configs/qwen36_27b_qlora_48k_openhands_segments_v2.yaml", |
| "base_model": "Qwen/Qwen3.6-27B", |
| "base_model_config": "Qwen/Qwen3.6-27B", |
| "batch_size": 8, |
| "bf16": true, |
| "capabilities": { |
| "bf16": true, |
| "compute_capability": "sm_80", |
| "fp8": false, |
| "n_gpu": 1, |
| "n_node": 1, |
| "tf32": true |
| }, |
| "context_parallel_size": 1, |
| "cut_cross_entropy": true, |
| "dataloader_num_workers": 2, |
| "dataloader_prefetch_factor": 2, |
| "dataset_num_proc": 12, |
| "dataset_prepared_path": "/content/outputs/last_run_prepared_48k_openhands_segments_v2", |
| "datasets": [ |
| { |
| "message_property_mappings": { |
| "content": "content", |
| "role": "role" |
| }, |
| "path": "/content/data/segments.train.jsonl", |
| "trust_remote_code": false, |
| "type": "input_output" |
| } |
| ], |
| "ddp": false, |
| "device": "cuda:0", |
| "device_map": "auto", |
| "dion_rank_fraction": 1.0, |
| "dion_rank_multiple_of": 1, |
| "eaft_alpha": 1.0, |
| "eaft_k": 20, |
| "embeddings_skip_upcast": true, |
| "env_capabilities": { |
| "torch_version": "2.10.0" |
| }, |
| "eot_tokens": [ |
| "<|im_end|>" |
| ], |
| "eval_batch_size": 1, |
| "eval_causal_lm_metrics": [ |
| "sacrebleu", |
| "comet", |
| "ter", |
| "chrf" |
| ], |
| "eval_max_new_tokens": 128, |
| "eval_sample_packing": true, |
| "eval_steps": 0.5, |
| "eval_table_size": 0, |
| "evals_per_epoch": 1, |
| "experimental_skip_move_to_device": true, |
| "flash_attention": false, |
| "fp16": false, |
| "generate_samples": false, |
| "generation_do_sample": true, |
| "generation_max_new_tokens": 50, |
| "generation_prompt_ratio": 0.5, |
| "generation_temperature": 0.7, |
| "gradient_accumulation_steps": 8, |
| "gradient_checkpointing": true, |
| "gradient_checkpointing_kwargs": { |
| "use_reentrant": false |
| }, |
| "greater_is_better": false, |
| "include_tkps": true, |
| "is_multimodal": true, |
| "layer_offloading": false, |
| "learning_rate": 0.0002, |
| "lisa_layers_attribute": "model.layers", |
| "load_best_model_at_end": true, |
| "load_in_4bit": false, |
| "load_in_8bit": false, |
| "local_rank": 0, |
| "logging_steps": 1, |
| "lora_alpha": 128, |
| "lora_dropout": 0.0, |
| "lora_mlp_kernel": true, |
| "lora_model_dir": "/content/outputs/qwen36-27b-qlora-48k-openhands", |
| "lora_o_kernel": true, |
| "lora_qkv_kernel": true, |
| "lora_r": 64, |
| "lora_target_modules": [ |
| "q_proj", |
| "k_proj", |
| "v_proj", |
| "o_proj", |
| "gate_proj", |
| "up_proj", |
| "down_proj" |
| ], |
| "loraplus_lr_embedding": 1e-06, |
| "lr_scheduler": "cosine", |
| "mean_resizing_embeddings": false, |
| "merge_lora": true, |
| "merge_method": "memory_efficient", |
| "metric_for_best_model": "eval_loss", |
| "micro_batch_size": 1, |
| "model_config_type": "qwen3_5", |
| "model_config_type_text": "qwen3_5_text", |
| "num_epochs": 2.0, |
| "num_generation_samples": 3, |
| "optimizer": "paged_adamw_8bit", |
| "otel_metrics_host": "localhost", |
| "otel_metrics_port": 8000, |
| "output_dir": "/content/outputs/qwen36-27b-qlora-48k-openhands/merged", |
| "pad_to_sequence_len": false, |
| "plugins": [ |
| "axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin" |
| ], |
| "pretrain_multipack_attn": true, |
| "processor_config": "Qwen/Qwen3.6-27B", |
| "profiler_steps_start": 0, |
| "qlora_sharded_model_loading": false, |
| "quantize_moe_experts": false, |
| "ray_num_workers": 1, |
| "resources_per_worker": { |
| "GPU": 1 |
| }, |
| "sample_packing": true, |
| "sample_packing_bin_size": 200, |
| "sample_packing_group_size": 100000, |
| "save_only_model": false, |
| "save_safetensors": true, |
| "save_steps": 0.25, |
| "save_total_limit": 6, |
| "saves_per_epoch": 2, |
| "seed": 42, |
| "sequence_len": 49152, |
| "shuffle_before_merging_datasets": false, |
| "shuffle_merged_datasets": true, |
| "skip_prepare_dataset": false, |
| "special_tokens": { |
| "eos_token": "<|im_end|>", |
| "pad_token": "<|endoftext|>" |
| }, |
| "streaming_multipack_buffer_size": 10000, |
| "strict": false, |
| "tensor_parallel_size": 1, |
| "test_datasets": [ |
| { |
| "message_property_mappings": { |
| "content": "content", |
| "role": "role" |
| }, |
| "path": "/content/data/segments.val.jsonl", |
| "trust_remote_code": false, |
| "type": "input_output" |
| } |
| ], |
| "tf32": true, |
| "tiled_mlp": true, |
| "tiled_mlp_use_original_mlp": true, |
| "tokenizer_config": "Qwen/Qwen3.6-27B", |
| "tokenizer_save_jinja_files": true, |
| "torch_dtype": "torch.bfloat16", |
| "train_on_inputs": false, |
| "trl": { |
| "async_prefetch": false, |
| "log_completions": false, |
| "mask_truncated_completions": false, |
| "ref_model_mixup_alpha": 0.9, |
| "ref_model_sync_steps": 64, |
| "replay_buffer_size": 0, |
| "replay_recompute_logps": true, |
| "reroll_max_groups": 1, |
| "reroll_start_fraction": 1.0, |
| "reward_num_workers": 1, |
| "scale_rewards": true, |
| "skip_zero_advantage_batches": true, |
| "sync_ref_model": false, |
| "use_data_producer": false, |
| "use_vllm": false, |
| "vllm_lora_sync": false, |
| "vllm_server_host": "0.0.0.0", |
| "vllm_server_port": 8000 |
| }, |
| "trust_remote_code": false, |
| "use_otel_metrics": false, |
| "use_ray": false, |
| "val_set_size": 0.0, |
| "vllm": { |
| "device": "auto", |
| "dtype": "auto", |
| "gpu_memory_utilization": 0.9, |
| "host": "0.0.0.0", |
| "port": 8000 |
| }, |
| "warmup_ratio": 0.03, |
| "weight_decay": 0.0, |
| "world_size": 1 |
| } |
| [2026-05-20 01:49:25,541] [DEBUG] [axolotl.cli.merge_lora.do_merge_lora:32] [PID:100066] Using memory-efficient LoRA merging method... |
| [2026-05-20 01:49:25,541] [DEBUG] [axolotl.cli.merge_lora._do_merge_lora_efficient:79] [PID:100066] Using memory-efficient LoRA merging method... |
|
Fetching 29 files: 0%| | 0/29 [00:00<?, ?it/s]
Fetching 29 files: 3%|▎ | 1/29 [00:00<00:09, 2.89it/s]
Fetching 29 files: 10%|█ | 3/29 [00:00<00:04, 6.03it/s]
Fetching 29 files: 28%|██▊ | 8/29 [00:00<00:01, 11.62it/s]
Fetching 29 files: 100%|██████████| 29/29 [00:00<00:00, 34.88it/s] |
| [2026-05-20 01:49:26,506] [DEBUG] [axolotl.cli.utils.lora_merge.merge_lora_sharded_efficient:838] [PID:100066] LoRA scale factor: 2.0 (rslora=False) |
| [2026-05-20 01:49:26,506] [INFO] [axolotl.cli.utils.lora_merge.merge_lora_sharded_efficient:841] [PID:100066] NF4 simulation enabled: base weights will undergo quantize→dequantize roundtrip before LoRA merge to match QLoRA training dynamics |
| [2026-05-20 01:49:26,506] [DEBUG] [axolotl.cli.utils.lora_merge.merge_lora_sharded_efficient:854] [PID:100066] Loading LoRA weights from /content/outputs/qwen36-27b-qlora-48k-openhands/adapter_model.safetensors |
| [2026-05-20 01:49:26,521] [DEBUG] [axolotl.cli.utils.lora_merge.merge_lora_sharded_efficient:860] [PID:100066] Keeping LoRA weights on CPU; will move per-tensor during merge |
| [2026-05-20 01:49:26,522] [DEBUG] [axolotl.cli.utils.lora_merge.merge_lora_sharded_efficient:866] [PID:100066] Found 15 model shards in /root/.cache/huggingface/hub/models--Qwen--Qwen3.6-27B/snapshots/6a9e13bd6fc8f0983b9b99948120bc37f49c13e9 |
| [2026-05-20 01:49:26,522] [INFO] [axolotl.cli.utils.lora_merge.copy_non_model_files:303] [PID:100066] Copying non-model files to output directory... |
| [2026-05-20 01:49:26,522] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying generation_config.json to output |
| [2026-05-20 01:49:26,523] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying preprocessor_config.json to output |
| [2026-05-20 01:49:26,523] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying config.json to output |
| [2026-05-20 01:49:26,523] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying vocab.json to output |
| [2026-05-20 01:49:26,527] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying merges.txt to output |
| [2026-05-20 01:49:26,529] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying chat_template.jinja to output |
| [2026-05-20 01:49:26,529] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying LICENSE to output |
| [2026-05-20 01:49:26,529] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying README.md to output |
| [2026-05-20 01:49:26,530] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying tokenizer_config.json to output |
| [2026-05-20 01:49:26,530] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying video_preprocessor_config.json to output |
| [2026-05-20 01:49:26,530] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying .gitattributes to output |
| [2026-05-20 01:49:26,530] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying tokenizer.json to output |
| [2026-05-20 01:49:26,537] [DEBUG] [axolotl.cli.utils.lora_merge.copy_non_model_files:324] [PID:100066] Copying configuration.json to output |
|
Merging shards: 0%| | 0/15 [00:00<?, ?it/s][2026-05-20 01:49:30,574] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.0.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:49:31,779] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.0.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:32,869] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.0.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:33,950] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.1.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:49:35,041] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.1.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:36,134] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.1.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:37,224] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.10.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:38,310] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.10.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
|
Merging shards: 7%|▋ | 1/15 [00:19<04:30, 19.32s/it][2026-05-20 01:49:46,109] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.10.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:49:47,226] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.11.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:49:48,352] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.11.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:49,461] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.11.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:50,556] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.12.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:49:51,650] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.12.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:52,741] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.12.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:53,831] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.13.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:49:54,918] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.13.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:56,005] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.13.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:57,089] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.14.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:49:58,180] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.14.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:49:59,269] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.14.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:00,364] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.15.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:01,466] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.15.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:02,549] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.15.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:03,630] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.16.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:04,715] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.16.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:05,795] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.16.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:06,880] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.17.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:07,967] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.17.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:09,053] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.17.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
|
Merging shards: 13%|█▎ | 2/15 [00:50<05:39, 26.12s/it][2026-05-20 01:50:16,969] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.18.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:18,072] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.18.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:19,184] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.18.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:20,272] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.19.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:21,363] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.19.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:22,457] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.19.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:23,553] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.2.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:24,662] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.2.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:25,752] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.2.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:26,842] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.20.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:27,935] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.20.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:29,014] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.20.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:30,086] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.21.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:31,166] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.21.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:32,242] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.21.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:33,319] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.22.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:34,401] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.22.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:35,479] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.22.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:36,557] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.23.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:37,640] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.23.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:38,729] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.23.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:39,822] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.24.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
|
Merging shards: 20%|██ | 3/15 [01:20<05:38, 28.19s/it][2026-05-20 01:50:47,623] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.24.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:48,784] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.24.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:49,890] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.25.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:51,000] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.25.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:52,115] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.25.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:53,213] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.26.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:54,295] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.26.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:55,375] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.26.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:56,452] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.27.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:50:57,533] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.27.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:58,614] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.27.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:50:59,695] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.28.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:00,776] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.28.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:01,857] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.28.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:02,938] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.29.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:04,025] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.29.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:05,112] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.29.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:06,191] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.3.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:07,266] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.3.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:08,339] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.3.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:09,427] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.30.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:10,512] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.30.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
|
Merging shards: 27%|██▋ | 4/15 [01:51<05:21, 29.21s/it][2026-05-20 01:51:18,702] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.30.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:20,571] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.31.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:22,352] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.31.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:24,105] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.31.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:25,865] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.4.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:27,589] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.4.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:29,276] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.4.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:30,451] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.5.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:31,547] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.5.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:32,637] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.5.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:33,727] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.6.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:34,812] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.6.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:35,889] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.6.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:36,972] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.7.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:38,056] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.7.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:39,137] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.7.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:40,218] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.8.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:41,304] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.8.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:42,394] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.8.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:43,487] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.9.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:44,569] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.9.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:51:45,650] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.9.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
|
Merging shards: 33%|███▎ | 5/15 [02:26<05:12, 31.27s/it][2026-05-20 01:51:57,612] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.11.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:51:57,726] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.11.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:51:58,368] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.11.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:51:59,640] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.11.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:01,880] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.15.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:01,939] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.15.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:52:02,326] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.15.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:52:03,092] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.15.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:04,122] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.19.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:04,176] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.19.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:52:04,558] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.19.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:52:05,317] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.19.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:06,017] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.23.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:06,078] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.23.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:52:06,851] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.23.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:06,904] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.27.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:06,954] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.27.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:52:07,726] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.27.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:07,785] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.3.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:07,835] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.3.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:52:08,604] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.3.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:08,662] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.31.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:08,722] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.31.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:52:09,489] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.31.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:09,546] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.7.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:52:09,604] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.7.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:52:10,364] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.7.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
|
Merging shards: 40%|████ | 6/15 [02:50<04:18, 28.67s/it][2026-05-20 01:52:19,251] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.23.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:52:23,596] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.27.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:52:26,526] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.3.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:52:28,239] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.31.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:52:31,181] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.7.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
|
Merging shards: 47%|████▋ | 7/15 [03:12<03:33, 26.73s/it][2026-05-20 01:52:53,216] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.32.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:52:54,610] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.32.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:52:55,986] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.32.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:52:57,503] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.33.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:52:59,099] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.33.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:00,689] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.33.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
|
Merging shards: 53%|█████▎ | 8/15 [03:43<03:14, 27.86s/it][2026-05-20 01:53:10,112] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.34.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:11,565] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.34.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:13,012] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.34.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:14,470] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.35.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:15,971] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.35.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:17,579] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.35.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:19,211] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.36.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:20,807] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.36.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:22,411] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.36.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:23,583] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.37.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:24,614] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.37.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:25,614] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.37.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:26,611] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.38.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:27,601] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.38.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:28,587] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.38.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:29,573] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.39.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:30,567] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.39.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:31,587] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.39.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:32,573] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.40.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:33,572] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.40.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:34,571] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.40.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:35,651] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.41.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
|
Merging shards: 60%|██████ | 9/15 [04:16<02:57, 29.55s/it][2026-05-20 01:53:43,721] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.41.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:45,434] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.41.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:47,095] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.42.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:48,741] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.42.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:50,451] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.42.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:52,070] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.43.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:53,675] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.43.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:55,306] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.43.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:56,928] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.44.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:53:58,537] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.44.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:53:59,928] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.44.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:01,551] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.45.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:03,148] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.45.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:04,775] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.45.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:06,404] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.46.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:08,025] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.46.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:09,682] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.46.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:11,389] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.47.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:13,046] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.47.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:14,772] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.47.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:16,389] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.48.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:17,942] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.48.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
|
Merging shards: 67%|██████▋ | 10/15 [04:59<02:47, 33.57s/it][2026-05-20 01:54:26,136] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.48.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:27,763] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.49.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:29,216] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.49.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:30,689] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.49.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:32,088] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.50.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:33,705] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.50.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:35,360] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.50.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:37,071] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.51.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:38,770] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.51.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:40,467] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.51.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:42,176] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.52.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:43,871] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.52.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:45,584] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.52.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:47,303] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.53.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:49,025] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.53.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:50,713] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.53.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:52,408] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.54.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:54,108] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.54.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:55,818] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.54.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:54:57,554] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.55.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:54:59,255] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.55.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:00,957] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.55.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
|
Merging shards: 73%|███████▎ | 11/15 [05:42<02:26, 36.52s/it][2026-05-20 01:55:09,431] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.56.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:55:10,856] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.56.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:12,307] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.56.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:13,787] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.57.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:55:15,462] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.57.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:17,080] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.57.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:18,662] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.58.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:55:20,256] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.58.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:21,837] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.58.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:23,396] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.59.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:55:24,980] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.59.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:26,557] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.59.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:28,166] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.60.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:55:29,803] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.60.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:31,403] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.60.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:32,987] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.61.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:55:34,601] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.61.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:36,207] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.61.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:37,769] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.62.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:55:39,379] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.62.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:40,968] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.62.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:55:42,599] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.63.mlp.gate_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
|
Merging shards: 80%|████████ | 12/15 [06:25<01:55, 38.60s/it][2026-05-20 01:55:55,553] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.35.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:55:55,666] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.35.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:55:56,184] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.35.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:55:57,406] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.35.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:00,852] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.39.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:00,962] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.39.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:56:01,582] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.39.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:56:02,814] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.39.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:03,746] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.43.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:03,848] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.43.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:56:05,081] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.43.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:05,183] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.47.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:05,282] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.47.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:56:06,503] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.47.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:06,607] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.51.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:06,708] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.51.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:56:07,930] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.51.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:08,031] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.55.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:08,132] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.55.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:56:09,345] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.55.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:09,446] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.59.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:09,549] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.59.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:56:10,765] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.59.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:10,866] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.63.mlp.down_proj.weight: torch.Size([64, 17408]), torch.Size([5120, 64]) |
| [2026-05-20 01:56:12,583] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.63.mlp.up_proj.weight: torch.Size([64, 5120]), torch.Size([17408, 64]) |
| [2026-05-20 01:56:14,323] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.63.self_attn.k_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
| [2026-05-20 01:56:14,421] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.63.self_attn.q_proj.weight: torch.Size([64, 5120]), torch.Size([12288, 64]) |
| [2026-05-20 01:56:15,653] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.63.self_attn.v_proj.weight: torch.Size([64, 5120]), torch.Size([1024, 64]) |
|
Merging shards: 87%|████████▋ | 13/15 [07:00<01:14, 37.42s/it][2026-05-20 01:56:29,999] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.43.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:56:33,899] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.47.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:56:37,967] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.51.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:56:42,025] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.55.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
| [2026-05-20 01:56:46,082] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.59.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
|
Merging shards: 93%|█████████▎| 14/15 [07:28<00:34, 34.53s/it][2026-05-20 01:56:56,987] [DEBUG] [axolotl.cli.utils.lora_merge._merge_tensor_with_lora:411] [PID:100066] Merging LoRA for model.language_model.layers.63.self_attn.o_proj.weight: torch.Size([64, 6144]), torch.Size([5120, 64]) |
|
Merging shards: 100%|██████████| 15/15 [07:31<00:00, 25.28s/it]
Merging shards: 100%|██████████| 15/15 [07:31<00:00, 30.13s/it] |
| [2026-05-20 01:56:58,522] [DEBUG] [axolotl.cli.utils.lora_merge.merge_lora_sharded_efficient:974] [PID:100066] Wrote weight-map index: model.safetensors.index.json |
| [2026-05-20 01:56:58,522] [INFO] [axolotl.cli.utils.lora_merge.merge_lora_sharded_efficient:982] [PID:100066] Applied LoRA to 256/1199 tensors |
| [2026-05-20 01:56:58,564] [DEBUG] [axolotl.cli.merge_lora._do_merge_lora_efficient:120] [PID:100066] Memory-efficient LoRA merge completed successfully! |
|
|
