Cartinoe5930/UST-1.5B-SFT-stage-1-depre

See axolotl config

axolotl version: 0.8.0.dev0

base_model: Qwen/Qwen2.5-1.5B-Instruct
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
trust_remote_code: false

load_in_8bit: false
load_in_4bit: false
strict: false

output_dir: ./outputs/out
chat_template: qwen_25
datasets:
  - path: Cartinoe5930/ust_sft_stage1
    type: chat_template
    # chat_template: tokenizer_default
    field_messages: messages
    message_field_role: role
    message_field_content: content
    roles:
      user:
        - user
      assistant:
        - assistant

dataset_prepared_path: last_run_prepared
val_set_size: 0.005
output_dir: ./outputs/out
eval_sample_packing: False

sequence_len: 8192
sample_packing: False
pad_to_sequence_len: False

wandb_project: hrm8k
wandb_entity:
wandb_watch: 
wandb_name: UST-1.5B-SFT-stage-1
hub_model_id: Cartinoe5930/UST-1.5B-SFT-stage-1

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

gradient_accumulation_steps: 8
micro_batch_size: 8
eval_batch_size: 4
num_epochs: 1
optimizer: paged_adamw_8bit
lr_scheduler: cosine
learning_rate: 1e-5

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

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

warmup_ratio: 0.05
evals_per_epoch: 1
eval_max_new_tokens: 128
eval_table_size:
saves_per_epoch: 1
debug:
deepspeed: deepspeed_configs/zero1.json
weight_decay: 0.01
fsdp:
fsdp_config:
special_tokens:

UST-1.5B-SFT-stage-1

This model is a fine-tuned version of Qwen/Qwen2.5-1.5B-Instruct on the Cartinoe5930/ust_sft_stage1 dataset. It achieves the following results on the evaluation set:

• Loss: 0.4671

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: 8
• eval_batch_size: 4
• seed: 42
• distributed_type: multi-GPU
• num_devices: 2
• gradient_accumulation_steps: 8
• total_train_batch_size: 128
• total_eval_batch_size: 8
• optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT 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: 9
• num_epochs: 1.0

Training results

Training Loss	Epoch	Step	Validation Loss
0.4262	0.9981	194	0.4671

Framework versions

• Transformers 4.50.0
• Pytorch 2.5.1+cu121
• Datasets 3.4.1
• Tokenizers 0.21.1