See axolotl config

axolotl version: 0.13.0.dev0

# === Model Configuration ===
base_model: Qwen/Qwen3-4B-Instruct-2507
load_in_8bit: false
load_in_4bit: false

# === Training Setup ===
num_epochs: 2
micro_batch_size: 1
gradient_accumulation_steps: 4
sequence_len: 2048
sample_packing: true
pad_to_sequence_len: true

# === Hyperparameter Configuration ===
optimizer: paged_ademamix_8bit
learning_rate: 1e-5
lr_scheduler: constant
weight_decay: 0.01
warmup_ratio: 0.05
cosine_min_lr_ratio: 0.1

# === Data Configuration ===
datasets:
  - path: WokeAI/polititune-tankie-warmup
    type: chat_template
    split: train
chat_template: tokenizer_default

dataset_prepared_path: last_run_prepared

# === Hardware Optimization ===
gradient_checkpointing: offload

# === Wandb Tracking ===
wandb_project: polititune-q34b-warmup

# === Checkpointing ===
saves_per_epoch: 2
save_only_model: true

# === Advanced Settings ===
output_dir: ./model-output
bf16: auto
flash_attention: true
train_on_inputs: false
group_by_length: false
logging_steps: 1
trust_remote_code: true
special_tokens:
  eos_token: <|im_end|>

model-output

This model is a fine-tuned version of Qwen/Qwen3-4B-Instruct-2507 on the WokeAI/polititune-tankie-warmup dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 4
total_train_batch_size: 4
optimizer: Use OptimizerNames.PAGED_ADEMAMIX_8BIT and the args are: No additional optimizer arguments
lr_scheduler_type: constant
lr_scheduler_warmup_steps: 2
training_steps: 22