Qwen3-1.7B-MT-v1

<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->

<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>

<details><summary>See axolotl config</summary>

axolotl version: 0.13.0.dev0

# ====== Model Configuration ======
base_model: Qwen/Qwen3-1.7B
load_in_8bit: false
load_in_4bit: true
strict: false

# LoRA
adapter: qlora
lora_r: 128
lora_alpha: 16
lora_target_linear: true
lora_qkv_kernel: true
lora_o_kernel: true
lora_mlp_kernel: true
embeddings_skip_upcast: true

# Integration
xformers_attention: true
# spa_attention: true
plugins:
  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_layer_norm: true
liger_fused_linear_cross_entropy: true
flash_attention: false

# DeepSpeed
# deepspeed: "/kaggle/working/axolotl/deepspeed_configs/zero2_torch_compile.json"

# ====== Hyperparameter Configuration ======
sample_packing: true
learning_rate: 2e-4
sequence_len: 2048
micro_batch_size: 16
gradient_accumulation_steps: 1
gradient_checkpointing: true
gradient_checkpointing_kwargs:
  use_reentrant: false

optimizer: adamw_8bit
lr_scheduler: cosine_with_min_lr
lr_scheduler_kwargs: {"min_lr_rate": 0.1}

warmup_ratio: 0.01
# warmup_steps: 0.01
weight_decay: 0.01
fp16: true

max_grad_norm: 1.0
num_epochs: 1
save_total_limit: 2
saves_per_epoch: 1
logging_steps: 1

output_dir: "/kaggle/working/outputs/qwen-sft-mt"
chat_template: qwen3

# Dataset
datasets:
  - path: Tung177/high_quality_MT
    type: chat_template
    split: "train"
    roles_to_train: ["assistant"]
dataset_prepared_path: last_run_prepared

dataloader_prefetch_factor: 8
dataloader_num_workers: 2
dataloader_pin_memory: true

# ====== Tracking ======
wandb_project: MT-V1
wandb_name: qwen3-1.7B-mt
wandb_log_model: "false"

</details><br>

kaggle/working/outputs/qwen-sft-mt

This model is a fine-tuned version of Qwen/Qwen3-1.7B on the Tung177/high_quality_MT 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: 0.0002
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• total_train_batch_size: 32
• total_eval_batch_size: 32
• optimizer: Use OptimizerNames.ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine_with_min_lr
• lr_scheduler_warmup_steps: 11
• training_steps: 1108
• mixed_precision_training: Native AMP

Training results

Framework versions

• PEFT 0.17.1
• Transformers 4.57.1
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.22.1