diff --git "a/debug.log" "b/debug.log"
new file mode 100644--- /dev/null
+++ "b/debug.log"
@@ -0,0 +1,11259 @@
+[2026-04-15 15:00:45,120] [DEBUG] [axolotl.utils.config.resolve_dtype:66] [PID:2788] bf16 support detected, enabling for this configuration.
+[2026-04-15 15:00:45,223] [DEBUG] [axolotl.utils.config.log_gpu_memory_usage:127] [PID:2788] baseline 0.000GB ()
+[2026-04-15 15:00:45,223] [INFO] [axolotl.cli.config.load_cfg:248] [PID:2788] config:
+{
+ "activation_offloading": false,
+ "adapter": "lora",
+ "axolotl_config_path": "config.yaml",
+ "base_model": "Qwen/Qwen2.5-Coder-3B-Instruct",
+ "base_model_config": "Qwen/Qwen2.5-Coder-3B-Instruct",
+ "batch_size": 8,
+ "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": 24,
+ "datasets": [
+ {
+ "chat_template": "tokenizer_default",
+ "ds_type": "json",
+ "field_messages": "messages",
+ "message_property_mappings": {
+ "content": "content",
+ "role": "role"
+ },
+ "path": "dria_pythonic_fc_chatml.jsonl",
+ "trust_remote_code": false,
+ "type": "chat_template"
+ }
+ ],
+ "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": 1,
+ "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": 8,
+ "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": 32,
+ "lora_dropout": 0.05,
+ "lora_r": 16,
+ "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": 1,
+ "model_config_type": "qwen2",
+ "num_epochs": 2.0,
+ "optimizer": "adamw_bnb_8bit",
+ "output_dir": "./outputs/Qwen2.5-Coder-3B-Instruct-coding-agent",
+ "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": 2048,
+ "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-Coder-3B-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
+}
+[2026-04-15 15:00:45,611] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:278] [PID:2788] EOS: 151645 / <|im_end|>
+[2026-04-15 15:00:45,616] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:2788] BOS: None / None
+[2026-04-15 15:00:45,617] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:2788] PAD: 151643 / <|endoftext|>
+[2026-04-15 15:00:45,618] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:2788] UNK: None / None
+[2026-04-15 15:00:45,618] [INFO] [axolotl.utils.data.shared.load_preprocessed_dataset:476] [PID:2788] Unable to find prepared dataset in last_run_prepared/8f7eb77e9223d8e0f9d67ceaf2fe66aa
+[2026-04-15 15:00:45,619] [INFO] [axolotl.utils.data.sft._load_raw_datasets:320] [PID:2788] Loading raw datasets...
+[2026-04-15 15:00:45,619] [WARNING] [axolotl.utils.data.sft._load_raw_datasets:322] [PID:2788] Processing datasets during training can lead to VRAM instability. Please pre-process your dataset using `axolotl preprocess path/to/config.yml`.
+
Generating train split: 0 examples [00:00, ? examples/s]
Generating train split: 4700 examples [00:00, 27428.78 examples/s]
Generating train split: 18689 examples [00:00, 65610.98 examples/s]
Generating train split: 31377 examples [00:00, 74707.26 examples/s]
Generating train split: 43579 examples [00:00, 81622.44 examples/s]
Generating train split: 52456 examples [00:00, 65689.00 examples/s]
Generating train split: 61915 examples [00:01, 48484.23 examples/s]
Generating train split: 70945 examples [00:01, 50576.27 examples/s]
Generating train split: 79532 examples [00:01, 53934.34 examples/s]
Generating train split: 81819 examples [00:01, 53985.94 examples/s]
+[2026-04-15 15:00:47,413] [INFO] [axolotl.utils.data.wrappers.get_dataset_wrapper:87] [PID:2788] Loading dataset: dria_pythonic_fc_chatml.jsonl with base_type: chat_template and prompt_style: None
+[2026-04-15 15:00:47,422] [INFO] [axolotl.prompt_strategies.chat_template.__call__:969] [PID:2788] Using chat template:
+---
+{%- if tools %}
+ {{- '<|im_start|>system\n' }}
+ {%- if messages[0]['role'] == 'system' %}
+ {{- messages[0]['content'] }}
+ {%- else %}
+ {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+ {%- endif %}
+ {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within XML tags:\n" }}
+ {%- for tool in tools %}
+ {{- "\n" }}
+ {{- tool | tojson }}
+ {%- endfor %}
+ {{- "\n\n\nFor each function call, return a json object with function name and arguments within XML tags:\n\n{\"name\": , \"arguments\": }\n<|im_end|>\n" }}
+{%- else %}
+ {%- if messages[0]['role'] == 'system' %}
+ {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+ {%- else %}
+ {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+ {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+ {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+ {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+ {%- elif message.role == "assistant" %}
+ {{- '<|im_start|>' + message.role }}
+ {%- if message.content %}
+ {{- '\n' + message.content }}
+ {%- endif %}
+ {%- for tool_call in message.tool_calls %}
+ {%- if tool_call.function is defined %}
+ {%- set tool_call = tool_call.function %}
+ {%- endif %}
+ {{- '\n\n{"name": "' }}
+ {{- tool_call.name }}
+ {{- '", "arguments": ' }}
+ {{- tool_call.arguments | tojson }}
+ {{- '}\n' }}
+ {%- endfor %}
+ {{- '<|im_end|>\n' }}
+ {%- elif message.role == "tool" %}
+ {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+ {{- '<|im_start|>user' }}
+ {%- endif %}
+ {{- '\n\n' }}
+ {{- message.content }}
+ {{- '\n' }}
+ {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+ {{- '<|im_end|>\n' }}
+ {%- endif %}
+ {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+ {{- '<|im_start|>assistant\n' }}
+{%- endif %}
+
+---
+
Tokenizing Prompts (num_proc=24): 0%| | 0/81819 [00:002048) (num_proc=24): 0%| | 0/81819 [00:002048) (num_proc=24): 1%|▍ | 1000/81819 [00:00<00:41, 1938.04 examples/s]
Dropping Long Sequences (>2048) (num_proc=24): 23%|██████▋ | 19000/81819 [00:00<00:01, 35010.37 examples/s]
Dropping Long Sequences (>2048) (num_proc=24): 35%|██████████▎ | 29000/81819 [00:00<00:01, 48529.89 examples/s]
Dropping Long Sequences (>2048) (num_proc=24): 51%|██████████████▉ | 42000/81819 [00:00<00:00, 62786.53 examples/s]
Dropping Long Sequences (>2048) (num_proc=24): 62%|██████████████████ | 51000/81819 [00:01<00:00, 68171.50 examples/s]
Dropping Long Sequences (>2048) (num_proc=24): 81%|███████████████████████▌ | 66637/81819 [00:01<00:00, 90094.56 examples/s]
Dropping Long Sequences (>2048) (num_proc=24): 96%|███████████████████████████▊ | 78365/81819 [00:01<00:00, 87181.77 examples/s]
Dropping Long Sequences (>2048) (num_proc=24): 100%|█████████████████████████████| 81819/81819 [00:01<00:00, 54016.24 examples/s]
+[2026-04-15 15:02:37,679] [WARNING] [axolotl.utils.data.utils.handle_long_seq_in_dataset:260] [PID:2788] Dropped 1351 samples from dataset
+
Saving the dataset (0/24 shards): 0%| | 0/80468 [00:00
+[2026-04-15 15:02:39,713] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:279] [PID:2788] BOS: None / None
+[2026-04-15 15:02:39,714] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:280] [PID:2788] PAD: 151643 / <|endoftext|>
+[2026-04-15 15:02:39,714] [DEBUG] [axolotl.loaders.tokenizer.load_tokenizer:281] [PID:2788] UNK: None / None
+[2026-04-15 15:02:39,714] [DEBUG] [axolotl.train.setup_model_and_tokenizer:74] [PID:2788] Loading model
+[2026-04-15 15:02:39,766] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_evaluation_loop:87] [PID:2788] Patched Trainer.evaluation_loop with nanmean loss calculation
+[2026-04-15 15:02:39,767] [DEBUG] [axolotl.monkeypatch.transformers.trainer_loss_calc.patch_maybe_log_save_evaluate:138] [PID:2788] Patched Trainer._maybe_log_save_evaluate with nanmean loss calculation
+
model.safetensors.index.json: 0.00B [00:00, ?B/s]
model.safetensors.index.json: 35.6kB [00:00, 170MB/s]
+
model-00001-of-00002.safetensors: 0%| | 0.00/4.96G [00:00