base_model: Qwen/Qwen2.5-Coder-7B-Instruct
model_type: Qwen2ForCausalLM
tokenizer_type: AutoTokenizer

load_in_8bit: false
load_in_4bit: false

datasets:
  - path: felixwangg/stage_1_combined
    type: chat_template
    split: train
test_datasets:
  - path: felixwangg/stage_1_combined
    type: chat_template
    split: validation
dataset_prepared_path: /home/tkwang/scratch/SecSteer-v2/axolotl-datasets/lora/Qwen2.5-Coder-7B/stage1-combined
val_set_size: 0
output_dir: /home/tkwang/scratch/SecSteer-v2/axolotl-outputs/lora/Qwen2.5-Coder-7B-stage1-combined
sequence_len: 4096
sample_packing: false
eval_sample_packing: false
pad_to_sequence_len: true

adapter: lora
lora_model_dir:
lora_r: 16
lora_alpha: 16
lora_dropout: 0.05
lora_target_linear: true
merge_lora: true

wandb_project: sft-primevul-sweep-ctx-0
wandb_entity: wtkuan
wandb_watch: "false"
wandb_name: Qwen2.5-Coder-7B-stage1-combined
wandb_log_model: "false"


gradient_accumulation_steps: 4
micro_batch_size: 4
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 4e-5

bf16: true
tf32: false

train_on_inputs: false
roles_to_train: ['assistant']

gradient_checkpointing: true
resume_from_checkpoint:
logging_steps: 1
flash_attention: true

num_epochs: 1
warmup_ratio: 0.1
early_stopping_patience: 1000
eval_steps: 2
save_steps: 2
save_total_limit: 1000
load_best_model_at_end: true

weight_decay: 0.02
special_tokens:

plugins:

home/tkwang/scratch/SecSteer-v2/axolotl-outputs/lora/Qwen2.5-Coder-7B-stage1-combined

This model is a fine-tuned version of Qwen/Qwen2.5-Coder-7B-Instruct on the felixwangg/stage_1_combined dataset. It achieves the following results on the evaluation set:

Loss: 0.8432
Ppl: 2.3238
Memory/max Active (gib): 38.24
Memory/max Allocated (gib): 38.24
Memory/device Reserved (gib): 51.33

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: 4e-05
train_batch_size: 4
eval_batch_size: 4
seed: 42
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 4
total_train_batch_size: 64
total_eval_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
training_steps: 6

Training results

Training Loss	Epoch	Step	Validation Loss	Ppl	Active (gib)	Allocated (gib)	Reserved (gib)
No log	0	0	0.8467	2.3320	37.91	37.91	41.87
0.8133	0.3636	2	0.8458	2.3298	38.24	38.24	51.33
0.7377	0.7273	4	0.8441	2.3259	38.24	38.24	51.33
0.8309	1.0	6	0.8432	2.3238	38.24	38.24	51.33

Framework versions

PEFT 0.18.1
Transformers 4.57.6
Pytorch 2.8.0+cu128
Datasets 4.5.0
Tokenizers 0.22.2

felixwangg/Qwen2.5-Coder-7B-stage1-combined-v2

home/tkwang/scratch/SecSteer-v2/axolotl-outputs/lora/Qwen2.5-Coder-7B-stage1-combined

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

felixwangg/
Qwen2.5-Coder-7B-stage1-combined-v2