[2025-10-22 16:34:24,607] [DEBUG] [axolotl.utils.config.resolve_dtype:66] [PID:2199] bf16 support detected, enabling for this configuration.
[2025-10-22 16:34:24,749] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:2199] baseline 0.000GB ()
[2025-10-22 16:34:24,749] [INFO] [axolotl.cli.config.load_cfg:248] [PID:2199] config:
{
  "activation_offloading": false,
  "adapter": "lora",
  "axolotl_config_path": "config.yaml",
  "base_model": "Qwen/Qwen2.5-7B-Instruct",
  "base_model_config": "Qwen/Qwen2.5-7B-Instruct",
  "batch_size": 16,
  "bf16": true,
  "capabilities": {
    "bf16": true,
    "compute_capability": "sm_90",
    "fp8": false,
    "n_gpu": 1,
    "n_node": 1
  },
  "context_parallel_size": 1,
  "dataloader_num_workers": 1,
  "dataloader_pin_memory": true,
  "dataloader_prefetch_factor": 256,
  "dataset_processes": 32,
  "datasets": [
    {
      "message_property_mappings": {
        "content": "content",
        "role": "role"
      },
      "path": "/workspace/fine-tuning/data/trump.json",
      "trust_remote_code": false,
      "type": "alpaca"
    }
  ],
  "ddp": false,
  "device": "cuda:0",
  "dion_rank_fraction": 1.0,
  "dion_rank_multiple_of": 1,
  "env_capabilities": {
    "torch_version": "2.7.1"
  },
  "eval_batch_size": 16,
  "eval_causal_lm_metrics": [
    "sacrebleu",
    "comet",
    "ter",
    "chrf"
  ],
  "eval_max_new_tokens": 128,
  "eval_table_size": 0,
  "experimental_skip_move_to_device": true,
  "fp16": false,
  "gradient_accumulation_steps": 1,
  "gradient_checkpointing": false,
  "include_tkps": true,
  "learning_rate": 0.0002,
  "lisa_layers_attribute": "model.layers",
  "load_best_model_at_end": false,
  "load_in_4bit": false,
  "load_in_8bit": true,
  "local_rank": 0,
  "lora_alpha": 16,
  "lora_dropout": 0.05,
  "lora_r": 8,
  "lora_target_modules": [
    "q_proj",
    "v_proj",
    "k_proj",
    "o_proj",
    "gate_proj",
    "down_proj",
    "up_proj"
  ],
  "loraplus_lr_embedding": 1e-06,
  "lr_scheduler": "cosine",
  "mean_resizing_embeddings": false,
  "micro_batch_size": 16,
  "model_config_type": "qwen2",
  "num_epochs": 1.0,
  "optimizer": "adamw_bnb_8bit",
  "output_dir": "./outputs/thoth_text_v2",
  "pretrain_multipack_attn": true,
  "profiler_steps_start": 0,
  "qlora_sharded_model_loading": false,
  "ray_num_workers": 1,
  "resources_per_worker": {
    "GPU": 1
  },
  "sample_packing_bin_size": 200,
  "sample_packing_group_size": 100000,
  "save_only_model": false,
  "save_safetensors": true,
  "sequence_len": 4096,
  "shuffle_before_merging_datasets": false,
  "shuffle_merged_datasets": true,
  "skip_prepare_dataset": false,
  "streaming_multipack_buffer_size": 10000,
  "strict": false,
  "tensor_parallel_size": 1,
  "tiled_mlp_use_original_mlp": true,
  "tokenizer_config": "Qwen/Qwen2.5-7B-Instruct",
  "tokenizer_save_jinja_files": true,
  "torch_dtype": "torch.bfloat16",
  "train_on_inputs": false,
  "trl": {
    "log_completions": false,
    "mask_truncated_completions": false,
    "ref_model_mixup_alpha": 0.9,
    "ref_model_sync_steps": 64,
    "scale_rewards": true,
    "sync_ref_model": false,
    "use_vllm": false,
    "vllm_server_host": "0.0.0.0",
    "vllm_server_port": 8000
  },
  "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
  },
  "weight_decay": 0.0,
  "world_size": 1
}
[2025-10-22 16:34:25,444] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:278] [PID:2199] EOS: 151645 / <|im_end|>
[2025-10-22 16:34:25,444] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:2199] BOS: None / None
[2025-10-22 16:34:25,444] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:2199] PAD: 151643 / <|endoftext|>
[2025-10-22 16:34:25,444] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:2199] UNK: None / None
[2025-10-22 16:34:25,445] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:476] [PID:2199] Unable to find prepared dataset in last_run_prepared/8eaeba3b90268710df560cd4dfe04e2d
[2025-10-22 16:34:25,445] [INFO] [axolotl.utils.data.sft._load_raw_datasets:320] [PID:2199] Loading raw datasets...
[2025-10-22 16:34:25,445] [WARNING] [axolotl.utils.data.sft._load_raw_datasets:322] [PID:2199] Processing datasets during training can lead to VRAM instability. Please pre-process your dataset using `axolotl preprocess path/to/config.yml`.
[2025-10-22 16:34:25,754] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:87] [PID:2199] Loading dataset: /workspace/fine-tuning/data/trump.json with base_type: alpaca and prompt_style: None
[2025-10-22 16:34:25,754] [WARNING] [datasets.arrow_dataset.map:3100] [PID:2199] num_proc must be <= 15. Reducing num_proc to 15 for dataset of size 15.

Tokenizing Prompts (num_proc=15):   0%|                                                                                    | 0/15 [00:00<?, ? examples/s]
Tokenizing Prompts (num_proc=15):   7%|█████                                                                       | 1/15 [00:00<00:08,  1.56 examples/s]
Tokenizing Prompts (num_proc=15):  13%|██████████▏                                                                 | 2/15 [00:00<00:04,  3.01 examples/s]
Tokenizing Prompts (num_proc=15):  27%|████████████████████▎                                                       | 4/15 [00:00<00:01,  5.98 examples/s]
Tokenizing Prompts (num_proc=15):  47%|███████████████████████████████████▍                                        | 7/15 [00:01<00:00,  9.94 examples/s]
Tokenizing Prompts (num_proc=15):  60%|█████████████████████████████████████████████▌                              | 9/15 [00:01<00:00, 11.64 examples/s]
Tokenizing Prompts (num_proc=15):  73%|██████████████████████████████████████████████████████▉                    | 11/15 [00:01<00:00, 12.43 examples/s]
Tokenizing Prompts (num_proc=15):  87%|█████████████████████████████████████████████████████████████████          | 13/15 [00:01<00:00, 12.76 examples/s]
Tokenizing Prompts (num_proc=15): 100%|███████████████████████████████████████████████████████████████████████████| 15/15 [00:01<00:00, 14.05 examples/s]
Tokenizing Prompts (num_proc=15): 100%|███████████████████████████████████████████████████████████████████████████| 15/15 [00:01<00:00,  8.67 examples/s]
[2025-10-22 16:34:27,600] [INFO] [axolotl.utils.data.utils.handle_long_seq_in_dataset:218] [PID:2199] min_input_len: 46
[2025-10-22 16:34:27,600] [INFO] [axolotl.utils.data.utils.handle_long_seq_in_dataset:220] [PID:2199] max_input_len: 87
[2025-10-22 16:34:27,601] [WARNING] [datasets.arrow_dataset.map:3100] [PID:2199] num_proc must be <= 15. Reducing num_proc to 15 for dataset of size 15.

Dropping Long Sequences (>4096) (num_proc=15):   0%|                                                                       | 0/15 [00:00<?, ? examples/s]
Dropping Long Sequences (>4096) (num_proc=15):   7%|████▏                                                          | 1/15 [00:00<00:03,  4.29 examples/s]
Dropping Long Sequences (>4096) (num_proc=15): 100%|██████████████████████████████████████████████████████████████| 15/15 [00:00<00:00, 34.65 examples/s]

Saving the dataset (0/1 shards):   0%|                                                                                     | 0/15 [00:00<?, ? examples/s]
Saving the dataset (1/1 shards): 100%|██████████████████████████████████████████████████████████████████████████| 15/15 [00:00<00:00, 1508.60 examples/s]
Saving the dataset (1/1 shards): 100%|██████████████████████████████████████████████████████████████████████████| 15/15 [00:00<00:00, 1468.01 examples/s]
[2025-10-22 16:34:28,163] [DEBUG] [axolotl.utils.trainer.calculate_total_num_steps:404] [PID:2199] total_num_tokens: 1_016
[2025-10-22 16:34:28,164] [DEBUG] [axolotl.utils.trainer.calculate_total_num_steps:422] [PID:2199] `total_supervised_tokens: 386`
[2025-10-22 16:34:28,164] [DEBUG] [axolotl.utils.trainer.calculate_total_num_steps:520] [PID:2199] total_num_steps: 1
[2025-10-22 16:34:28,164] [INFO] [axolotl.utils.data.sft._prepare_standard_dataset:121] [PID:2199] Maximum number of steps set at 1
[2025-10-22 16:34:28,208] [DEBUG] [axolotl.train.setup_model_and_tokenizer:65] [PID:2199] Loading tokenizer... Qwen/Qwen2.5-7B-Instruct
[2025-10-22 16:34:28,772] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:278] [PID:2199] EOS: 151645 / <|im_end|>
[2025-10-22 16:34:28,772] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:2199] BOS: None / None
[2025-10-22 16:34:28,772] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:2199] PAD: 151643 / <|endoftext|>
[2025-10-22 16:34:28,772] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:2199] UNK: None / None
[2025-10-22 16:34:28,772] [DEBUG] [axolotl.train.setup_model_and_tokenizer:74] [PID:2199] Loading model
[2025-10-22 16:34:28,893] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_evaluation_loop:87] [PID:2199] Patched Trainer.evaluation_loop with nanmean loss calculation
[2025-10-22 16:34:28,895] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_maybe_log_save_evaluate:138] [PID:2199] Patched Trainer._maybe_log_save_evaluate with nanmean loss calculation

model.safetensors.index.json: 0.00B [00:00, ?B/s]
model.safetensors.index.json: 27.8kB [00:00, 71.6MB/s]

model-00001-of-00004.safetensors:   0%|                                                                                      | 0.00/3.95G [00:00<?, ?B/s]
model-00001-of-00004.safetensors:   0%|                                                                          | 52.0k/3.95G [00:01<35:38:11, 30.8kB/s]
model-00001-of-00004.safetensors:   1%|█                                                                            | 52.6M/3.95G [00:01<01:34, 41.0MB/s]
model-00001-of-00004.safetensors:   2%|█▋                                                                           | 83.8M/3.95G [00:02<01:22, 46.6MB/s]
model-00001-of-00004.safetensors:   6%|████▎                                                                          | 218M/3.95G [00:02<00:23, 157MB/s]
model-00001-of-00004.safetensors:  11%|████████▉                                                                      | 446M/3.95G [00:02<00:09, 389MB/s]
model-00001-of-00004.safetensors:  14%|███████████▍                                                                   | 571M/3.95G [00:02<00:07, 458MB/s]
model-00001-of-00004.safetensors:  18%|██████████████▌                                                                | 728M/3.95G [00:02<00:06, 533MB/s]
model-00001-of-00004.safetensors:  21%|████████████████▋                                                              | 835M/3.95G [00:03<00:06, 500MB/s]
model-00001-of-00004.safetensors:  23%|██████████████████▎                                                            | 917M/3.95G [00:03<00:07, 413MB/s]
model-00001-of-00004.safetensors:  25%|███████████████████▉                                                           | 998M/3.95G [00:03<00:06, 445MB/s]
model-00001-of-00004.safetensors:  29%|██████████████████████▎                                                       | 1.13G/3.95G [00:03<00:05, 496MB/s]
model-00001-of-00004.safetensors:  32%|█████████████████████████▎                                                    | 1.28G/3.95G [00:04<00:04, 647MB/s]
model-00001-of-00004.safetensors:  36%|███████████████████████████▉                                                  | 1.41G/3.95G [00:04<00:03, 675MB/s]
model-00001-of-00004.safetensors:  38%|█████████████████████████████▉                                                | 1.51G/3.95G [00:04<00:03, 713MB/s]
model-00001-of-00004.safetensors:  42%|█████████████████████████████████                                             | 1.67G/3.95G [00:04<00:02, 892MB/s]
model-00001-of-00004.safetensors:  49%|█████████████████████████████████████▍                                       | 1.92G/3.95G [00:04<00:01, 1.19GB/s]
model-00001-of-00004.safetensors:  54%|█████████████████████████████████████████▏                                   | 2.11G/3.95G [00:04<00:01, 1.07GB/s]
model-00001-of-00004.safetensors:  60%|█████████████████████████████████████████████▊                               | 2.35G/3.95G [00:04<00:01, 1.21GB/s]
model-00001-of-00004.safetensors:  63%|████████████████████████████████████████████████▍                            | 2.48G/3.95G [00:05<00:01, 1.18GB/s]
model-00001-of-00004.safetensors:  67%|███████████████████████████████████████████████████▊                         | 2.65G/3.95G [00:05<00:01, 1.14GB/s]
model-00001-of-00004.safetensors:  78%|███████████████████████████████████████████████████████████▋                 | 3.06G/3.95G [00:05<00:00, 1.75GB/s]
model-00001-of-00004.safetensors:  85%|█████████████████████████████████████████████████████████████████▍           | 3.36G/3.95G [00:05<00:00, 1.88GB/s]
model-00001-of-00004.safetensors:  91%|██████████████████████████████████████████████████████████████████████▍      | 3.61G/3.95G [00:05<00:00, 1.99GB/s]
model-00001-of-00004.safetensors: 100%|█████████████████████████████████████████████████████████████████████████████| 3.95G/3.95G [00:05<00:00, 2.22GB/s]
model-00001-of-00004.safetensors: 100%|██████████████████████████████████████████████████████████████████████████████| 3.95G/3.95G [00:05<00:00, 697MB/s]

model-00002-of-00004.safetensors:   0%|                                                                                      | 0.00/3.86G [00:00<?, ?B/s]
model-00002-of-00004.safetensors:   0%|                                                                             | 3.45M/3.86G [00:01<32:49, 1.96MB/s]
model-00002-of-00004.safetensors:   3%|██▏                                                                           | 108M/3.86G [00:02<01:04, 58.4MB/s]
model-00002-of-00004.safetensors:   9%|██████▉                                                                        | 338M/3.86G [00:02<00:18, 190MB/s]
model-00002-of-00004.safetensors:  10%|███████▊                                                                       | 384M/3.86G [00:02<00:16, 206MB/s]
model-00002-of-00004.safetensors:  15%|███████████▊                                                                   | 576M/3.86G [00:02<00:08, 377MB/s]
model-00002-of-00004.safetensors:  18%|██████████████▍                                                                | 709M/3.86G [00:03<00:06, 496MB/s]
model-00002-of-00004.safetensors:  28%|█████████████████████▋                                                        | 1.08G/3.86G [00:03<00:02, 945MB/s]
model-00002-of-00004.safetensors:  33%|█████████████████████████▍                                                    | 1.26G/3.86G [00:03<00:02, 940MB/s]
model-00002-of-00004.safetensors:  37%|█████████████████████████████▏                                                | 1.45G/3.86G [00:03<00:02, 983MB/s]
model-00002-of-00004.safetensors:  42%|████████████████████████████████▋                                            | 1.64G/3.86G [00:03<00:02, 1.01GB/s]
model-00002-of-00004.safetensors:  48%|████████████████████████████████████▋                                        | 1.84G/3.86G [00:03<00:01, 1.11GB/s]
model-00002-of-00004.safetensors:  54%|█████████████████████████████████████████▎                                   | 2.07G/3.86G [00:03<00:01, 1.24GB/s]
model-00002-of-00004.safetensors:  60%|█████████████████████████████████████████████▊                               | 2.30G/3.86G [00:04<00:01, 1.42GB/s]
model-00002-of-00004.safetensors:  73%|███████████████████████████████████████████████████████▉                     | 2.81G/3.86G [00:04<00:00, 2.16GB/s]
model-00002-of-00004.safetensors:  80%|█████████████████████████████████████████████████████████████▏               | 3.07G/3.86G [00:04<00:00, 1.90GB/s]
model-00002-of-00004.safetensors:  86%|██████████████████████████████████████████████████████████████████▌          | 3.34G/3.86G [00:04<00:00, 1.68GB/s]
model-00002-of-00004.safetensors:  97%|██████████████████████████████████████████████████████████████████████████▋  | 3.75G/3.86G [00:04<00:00, 2.08GB/s]
model-00002-of-00004.safetensors: 100%|██████████████████████████████████████████████████████████████████████████████| 3.86G/3.86G [00:05<00:00, 703MB/s]

model-00003-of-00004.safetensors:   0%|                                                                                      | 0.00/3.86G [00:00<?, ?B/s]
model-00003-of-00004.safetensors:   0%|                                                                            | 1.37M/3.86G [00:01<1:18:05, 824kB/s]
model-00003-of-00004.safetensors:   1%|▉                                                                            | 50.0M/3.86G [00:02<02:08, 29.6MB/s]
model-00003-of-00004.safetensors:   6%|████▉                                                                          | 243M/3.86G [00:02<00:23, 152MB/s]
model-00003-of-00004.safetensors:  10%|███████▉                                                                       | 388M/3.86G [00:02<00:13, 254MB/s]
model-00003-of-00004.safetensors:  12%|█████████▍                                                                     | 463M/3.86G [00:03<00:14, 234MB/s]
model-00003-of-00004.safetensors:  15%|███████████▌                                                                   | 564M/3.86G [00:03<00:11, 279MB/s]
model-00003-of-00004.safetensors:  21%|████████████████▎                                                              | 797M/3.86G [00:03<00:06, 511MB/s]
model-00003-of-00004.safetensors:  28%|██████████████████████▏                                                       | 1.10G/3.86G [00:03<00:03, 806MB/s]
model-00003-of-00004.safetensors:  35%|███████████████████████████▏                                                 | 1.37G/3.86G [00:03<00:02, 1.00GB/s]
model-00003-of-00004.safetensors:  42%|████████████████████████████████▌                                            | 1.63G/3.86G [00:03<00:01, 1.27GB/s]
model-00003-of-00004.safetensors:  47%|████████████████████████████████████▍                                        | 1.83G/3.86G [00:04<00:01, 1.13GB/s]
model-00003-of-00004.safetensors:  68%|████████████████████████████████████████████████████▍                        | 2.63G/3.86G [00:04<00:00, 2.36GB/s]
model-00003-of-00004.safetensors:  77%|███████████████████████████████████████████████████████████                  | 2.97G/3.86G [00:04<00:00, 2.15GB/s]
model-00003-of-00004.safetensors:  85%|█████████████████████████████████████████████████████████████████▋           | 3.29G/3.86G [00:04<00:00, 1.95GB/s]
model-00003-of-00004.safetensors:  95%|█████████████████████████████████████████████████████████████████████████▎   | 3.68G/3.86G [00:04<00:00, 2.19GB/s]
model-00003-of-00004.safetensors: 100%|██████████████████████████████████████████████████████████████████████████████| 3.86G/3.86G [00:05<00:00, 704MB/s]

model-00004-of-00004.safetensors:   0%|                                                                                      | 0.00/3.56G [00:00<?, ?B/s]
model-00004-of-00004.safetensors:   0%|                                                                         | 11.9k/3.56G [00:01<134:42:23, 7.33kB/s]
model-00004-of-00004.safetensors:   0%|▎                                                                            | 12.1M/3.56G [00:02<07:39, 7.72MB/s]
model-00004-of-00004.safetensors:   3%|██                                                                           | 95.0M/3.56G [00:02<00:45, 75.9MB/s]
model-00004-of-00004.safetensors:   5%|███▋                                                                           | 166M/3.56G [00:02<00:24, 139MB/s]
model-00004-of-00004.safetensors:   9%|███████▏                                                                       | 323M/3.56G [00:02<00:10, 322MB/s]
model-00004-of-00004.safetensors:  12%|█████████▎                                                                     | 419M/3.56G [00:02<00:09, 327MB/s]
model-00004-of-00004.safetensors:  17%|█████████████▋                                                                 | 617M/3.56G [00:02<00:05, 557MB/s]
model-00004-of-00004.safetensors:  21%|████████████████▋                                                              | 750M/3.56G [00:03<00:05, 524MB/s]
model-00004-of-00004.safetensors:  26%|████████████████████▍                                                          | 917M/3.56G [00:03<00:04, 537MB/s]
model-00004-of-00004.safetensors:  30%|███████████████████████▍                                                      | 1.07G/3.56G [00:03<00:03, 649MB/s]
model-00004-of-00004.safetensors:  45%|██████████████████████████████████▉                                          | 1.61G/3.56G [00:03<00:01, 1.43GB/s]
model-00004-of-00004.safetensors:  56%|███████████████████████████████████████████▏                                 | 1.99G/3.56G [00:03<00:00, 1.65GB/s]
model-00004-of-00004.safetensors:  65%|██████████████████████████████████████████████████▎                          | 2.32G/3.56G [00:03<00:00, 1.90GB/s]
model-00004-of-00004.safetensors:  73%|███████████████████████████████████████████████████████▉                     | 2.58G/3.56G [00:04<00:00, 1.45GB/s]
model-00004-of-00004.safetensors:  81%|██████████████████████████████████████████████████████████████▌              | 2.89G/3.56G [00:04<00:00, 1.55GB/s]
model-00004-of-00004.safetensors:  87%|██████████████████████████████████████████████████████████████████▉          | 3.09G/3.56G [00:04<00:00, 1.48GB/s]
model-00004-of-00004.safetensors:  93%|███████████████████████████████████████████████████████████████████████▎     | 3.29G/3.56G [00:04<00:00, 1.34GB/s]
model-00004-of-00004.safetensors:  98%|███████████████████████████████████████████████████████████████████████████▌ | 3.49G/3.56G [00:04<00:00, 1.35GB/s]
model-00004-of-00004.safetensors: 100%|██████████████████████████████████████████████████████████████████████████████| 3.56G/3.56G [00:04<00:00, 727MB/s]

Loading checkpoint shards:   0%|                                                                                                   | 0/4 [00:00<?, ?it/s]
Loading checkpoint shards:  25%|██████████████████████▊                                                                    | 1/4 [00:04<00:12,  4.04s/it]
Loading checkpoint shards:  50%|█████████████████████████████████████████████▌                                             | 2/4 [00:09<00:09,  4.79s/it]
Loading checkpoint shards:  75%|████████████████████████████████████████████████████████████████████▎                      | 3/4 [00:14<00:05,  5.01s/it]
Loading checkpoint shards: 100%|███████████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:18<00:00,  4.46s/it]
Loading checkpoint shards: 100%|███████████████████████████████████████████████████████████████████████████████████████████| 4/4 [00:18<00:00,  4.56s/it]

generation_config.json:   0%|                                                                                                  | 0.00/243 [00:00<?, ?B/s]
generation_config.json: 100%|███████████████████████████████████████████████████████████████████████████████████████████| 243/243 [00:00<00:00, 2.84MB/s]
[2025-10-22 16:35:11,043] [INFO] [axolotl.loaders.model._prepare_model_for_quantization:863] [PID:2199] converting PEFT model w/ prepare_model_for_kbit_training
[2025-10-22 16:35:11,045] [INFO] [axolotl.loaders.model._configure_embedding_dtypes:345] [PID:2199] Converting modules to torch.bfloat16
[2025-10-22 16:35:11,047] [DEBUG] [axolotl.loaders.model.log_gpu_memory_usage:127] [PID:2199] Memory usage after model load 11.676GB (+11.676GB allocated, +13.172GB reserved)
trainable params: 20,185,088 || all params: 7,635,801,600 || trainable%: 0.2643
[2025-10-22 16:35:11,254] [DEBUG] [axolotl.loaders.model.log_gpu_memory_usage:127] [PID:2199] after adapters 8.565GB (+8.565GB allocated, +13.248GB reserved)
[2025-10-22 16:35:16,678] [INFO] [axolotl.train.save_initial_configs:398] [PID:2199] Pre-saving adapter config to ./outputs/thoth_text_v2...
[2025-10-22 16:35:16,678] [INFO] [axolotl.train.save_initial_configs:402] [PID:2199] Pre-saving tokenizer to ./outputs/thoth_text_v2...
[2025-10-22 16:35:16,788] [INFO] [axolotl.train.save_initial_configs:407] [PID:2199] Pre-saving model config to ./outputs/thoth_text_v2...
[2025-10-22 16:35:16,790] [INFO] [axolotl.train.execute_training:196] [PID:2199] Starting trainer...

  0%|                                                                                                                              | 0/1 [00:00<?, ?it/s][2025-10-22 16:35:17,741] [WARNING] [py.warnings._showwarnmsg:110] [PID:2199] /root/miniconda3/envs/py3.11/lib/python3.11/site-packages/bitsandbytes/autograd/_functions.py:186: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization
  warnings.warn(f"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization")


100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00,  1.18s/it]
                                                                                                                                                         
{'loss': 3.0765, 'grad_norm': 2.254563570022583, 'learning_rate': 0.0002, 'memory/max_active (GiB)': 30.16, 'memory/max_allocated (GiB)': 30.16, 'memory/device_reserved (GiB)': 30.95, 'tokens_per_second_per_gpu': 466.19, 'epoch': 1.0}

100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00,  1.18s/it][2025-10-22 16:35:18,399] [INFO] [axolotl.core.trainers.base._save:671] [PID:2199] Saving model checkpoint to ./outputs/thoth_text_v2/checkpoint-1

                                                                                                                                                         
{'train_runtime': 1.6797, 'train_samples_per_second': 9.525, 'train_steps_per_second': 0.595, 'train_loss': 3.0764801502227783, 'memory/max_active (GiB)': 8.67, 'memory/max_allocated (GiB)': 8.67, 'memory/device_reserved (GiB)': 30.95, 'epoch': 1.0}

100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00,  1.18s/it]
100%|██████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 1/1 [00:01<00:00,  1.68s/it]
[2025-10-22 16:35:18,999] [INFO] [axolotl.train.save_trained_model:218] [PID:2199] Training completed! Saving trained model to ./outputs/thoth_text_v2.
[2025-10-22 16:35:19,335] [INFO] [axolotl.train.save_trained_model:336] [PID:2199] Model successfully saved to ./outputs/thoth_text_v2