See axolotl config

axolotl version: 0.15.0.dev0

base_model: Qwen/Qwen2.5-1.5B-Instruct

load_in_8bit: false
load_in_4bit: false

datasets:
  - path: train_T2J.jsonl
    type: chat_template

dataset_prepared_path: preprocess
val_set_size: 0.01
output_dir: ./outputs

adapter: 
lora_model_dir:

sequence_len: 16384
sample_packing: false
eval_sample_packing: false
pad_to_sequence_len: false

plugins:
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_swiglu: true
liger_fused_linear_cross_entropy: true

wandb_project: FC-T2J
wandb_entity: 
wandb_watch:
wandb_name: 
wandb_log_model:
hub_model_id: amphora/FC-T2J-SFT-1_5B

gradient_accumulation_steps: 128
micro_batch_size: 2
num_epochs: 3
optimizer: adamw_torch_fused
lr_scheduler: cosine
learning_rate: 2e-5

bf16: auto
tf32: false

gradient_checkpointing:
resume_from_checkpoint:
logging_steps: 1
flash_attention: true

warmup_ratio: 0.05
weight_decay: 0.01
evals_per_epoch: 0
saves_per_epoch: 1

# fsdp_config:
#   fsdp_version: 2
#   fsdp_offload_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
#   fsdp_reshard_after_forward: true
#   fsdp_activation_checkpointing: true

FC-T2J-SFT-1_5B

This model is a fine-tuned version of Qwen/Qwen2.5-1.5B-Instruct on the train_T2J.jsonl 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: 2e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• gradient_accumulation_steps: 128
• total_train_batch_size: 256
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 87
• training_steps: 1751

Training results

Framework versions

• Transformers 5.0.0
• Pytorch 2.9.1+cu128
• Datasets 4.5.0
• Tokenizers 0.22.2