Tibogoss/Qwen3-8B-test

base_model: Qwen/Qwen3-8B
# Automatically upload checkpoint and final model to HF
# hub_model_id: username/custom_model_name

quantization_config:
  load_in_4bit: true
  bnb_4bit_quant_type: "nf4"
  bnb_4bit_compute_dtype: "bfloat16"
  bnb_4bit_use_double_quant: true
strict: false

datasets:
  - path: /workspace/outputs/training_data/
    ds_type: json
    data_files: 
        - knowledge_v2.json
    type: chat_template
dataset_prepared_path:
val_set_size: 0.05
output_dir: /workspace/outputs/FT_v2

sequence_len: 1024
sample_packing: true
eval_sample_packing: true


adapter: lora
lora_model_dir:
lora_r: 32
lora_alpha: 64
lora_dropout: 0.05
lora_target_linear: true

wandb_project:
wandb_entity:
wandb_watch:
wandb_name:
wandb_log_model:

gradient_accumulation_steps: 4
micro_batch_size: 1
num_epochs: 1
optimizer: adamw_torch_fused
lr_scheduler: cosine
learning_rate: 0.0002

bf16: auto
tf32: true

gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false
resume_from_checkpoint:
logging_steps: 1
flash_attention: true

warmup_ratio: 0.1
evals_per_epoch: 4
saves_per_epoch: 1
weight_decay: 0.0
#fsdp:
#  - full_shard
#  - auto_wrap
#fsdp_config:
#  fsdp_limit_all_gathers: true
#  fsdp_sync_module_states: true
#  fsdp_offload_params: true
#  fsdp_use_orig_params: false
#  fsdp_cpu_ram_efficient_loading: true
#  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
#  fsdp_transformer_layer_cls_to_wrap: Qwen3DecoderLayer
#  fsdp_state_dict_type: FULL_STATE_DICT
#  fsdp_sharding_strategy: FULL_SHARD
special_tokens:

# save_first_step: true  # uncomment this to validate checkpoint saving works with your config