metadata
library_name: peft
license: apache-2.0
base_model: unsloth/Qwen2.5-Coder-1.5B-Instruct
tags:
- axolotl
- generated_from_trainer
model-index:
- name: fb1ffda1-f53f-4f10-81b7-203461bcc737
results: []
See axolotl config
axolotl version: 0.4.1
adapter: lora
base_model: unsloth/Qwen2.5-Coder-1.5B-Instruct
bf16: true
chat_template: llama3
dataset_prepared_path: null
datasets:
- data_files:
- e3e040151a7be6d4_train_data.json
ds_type: json
format: custom
path: /workspace/input_data/e3e040151a7be6d4_train_data.json
type:
field_instruction: text
field_output: extracted
format: '{instruction}'
no_input_format: '{instruction}'
system_format: '{system}'
system_prompt: ''
debug: null
deepspeed: null
early_stopping_patience: 2
eval_max_new_tokens: 128
eval_steps: 100
eval_table_size: null
flash_attention: true
fp16: null
fsdp: null
fsdp_config: null
gradient_accumulation_steps: 8
gradient_checkpointing: true
group_by_length: false
hub_model_id: Alphatao/fb1ffda1-f53f-4f10-81b7-203461bcc737
hub_repo: null
hub_strategy: checkpoint
hub_token: null
learning_rate: 0.0002
load_best_model_at_end: true
load_in_4bit: false
load_in_8bit: false
local_rank: null
logging_steps: 1
lora_alpha: 32
lora_dropout: 0.05
lora_fan_in_fan_out: null
lora_model_dir: null
lora_r: 16
lora_target_linear: true
lora_target_modules:
- q_proj
- k_proj
- v_proj
lr_scheduler: cosine
max_grad_norm: 1.0
max_steps: 3060
micro_batch_size: 4
mlflow_experiment_name: /tmp/e3e040151a7be6d4_train_data.json
model_type: AutoModelForCausalLM
num_epochs: 2
optimizer: adamw_bnb_8bit
output_dir: miner_id_24
pad_to_sequence_len: true
resume_from_checkpoint: null
s2_attention: null
sample_packing: false
save_steps: 100
sequence_len: 1024
strict: false
tf32: true
tokenizer_type: AutoTokenizer
train_on_inputs: false
trust_remote_code: true
val_set_size: 0.013473239451800833
wandb_entity: null
wandb_mode: online
wandb_name: 7f20767d-f9c0-42d6-8ba1-0f6d9bdd9674
wandb_project: Gradients-On-Demand
wandb_run: your_name
wandb_runid: 7f20767d-f9c0-42d6-8ba1-0f6d9bdd9674
warmup_steps: 10
weight_decay: 0.0
xformers_attention: null
fb1ffda1-f53f-4f10-81b7-203461bcc737
This model is a fine-tuned version of unsloth/Qwen2.5-Coder-1.5B-Instruct on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0478
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: 4
- eval_batch_size: 4
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 32
- optimizer: Use OptimizerNames.ADAMW_BNB 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: 10
- training_steps: 3060
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 0.7826 | 0.0001 | 1 | 2.1896 |
| 0.0552 | 0.0087 | 100 | 0.0724 |
| 0.0327 | 0.0175 | 200 | 0.0615 |
| 0.0452 | 0.0262 | 300 | 0.0588 |
| 0.0337 | 0.0350 | 400 | 0.0561 |
| 0.0411 | 0.0437 | 500 | 0.0528 |
| 0.0244 | 0.0524 | 600 | 0.0535 |
| 0.0323 | 0.0612 | 700 | 0.0510 |
| 0.0277 | 0.0699 | 800 | 0.0505 |
| 0.0459 | 0.0787 | 900 | 0.0496 |
| 0.0495 | 0.0874 | 1000 | 0.0483 |
| 0.037 | 0.0961 | 1100 | 0.0482 |
| 0.031 | 0.1049 | 1200 | 0.0485 |
| 0.0415 | 0.1136 | 1300 | 0.0455 |
| 0.031 | 0.1224 | 1400 | 0.0452 |
| 0.026 | 0.1311 | 1500 | 0.0455 |
| 0.0365 | 0.1398 | 1600 | 0.0478 |
Framework versions
- PEFT 0.13.2
- Transformers 4.46.0
- Pytorch 2.5.0+cu124
- Datasets 3.0.1
- Tokenizers 0.20.1