|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
datasets: |
|
- berkeley-nest/Nectar |
|
pipeline_tag: text-generation |
|
base_model: Na0s/Llama-3.1-8B-Pruned-4-Layers_LoRA-PEFT |
|
--- |
|
|
|
<a href="https://ibb.co/NtQ3QfF"><img src="https://i.ibb.co/RYZSZtg/model.webp" alt="model" border="0" alt="Model-card-peft-lora-1.0" align="center">></a> alt="Model-card-peft-lora-1.0" align="center"> |
|
|
|
# Model Card for Na0s/Llama-3.1-8B-Pruned-4-Layers_LoRA-PEFT-1.0 |
|
|
|
## Model Details |
|
|
|
### Model Description |
|
- **Finetuned from model:[Na0s/Llama-3.1-8B-Pruned-4-Layers_LoRA-PEFT]** |
|
|
|
|
|
## Training Details |
|
# Parameters used for Na0s/Llama-3.1-8B-Pruned-4-Layers_LoRA-PEFT-1.0 |
|
|
|
|
|
model = FastLanguageModel.get_peft_model( |
|
|
|
model, |
|
r = 16, |
|
target_modules = ["q_proj", "k_proj", "v_proj", "o_proj", |
|
"gate_proj", "up_proj", "down_proj",], |
|
|
|
lora_alpha = 16, |
|
lora_dropout = 0.05, |
|
bias = "none", |
|
|
|
use_gradient_checkpointing = "unsloth", |
|
random_state = 3407, |
|
use_rslora = False, |
|
loftq_config = None, |
|
) |
|
|
|
trainer = SFTTrainer( |
|
|
|
model = model, |
|
tokenizer = tokenizer, |
|
train_dataset = dataset, |
|
dataset_text_field = "completion", |
|
|
|
max_seq_length = max_seq_length, |
|
dataset_num_proc = 2, |
|
packing = False, |
|
|
|
args = TrainingArguments( |
|
per_device_train_batch_size = 6, |
|
gradient_accumulation_steps = 4, |
|
warmup_steps = 5, |
|
max_steps=5000, |
|
learning_rate = 2e-4, |
|
fp16 = not is_bfloat16_supported(), |
|
|
|
bf16 = is_bfloat16_supported(), |
|
logging_steps = 1, |
|
optim = "adamw_8bit", |
|
weight_decay = 0.01, |
|
lr_scheduler_type = "linear", |
|
seed = 3407, |
|
output_dir = "outputs_2", |
|
push_to_hub=True, |
|
hub_always_push=True, |
|
), |
|
) |
|
|
|
Dataset: Berkeley-nest/Nectar |
|
|
|
### Training Data |
|
|
|
<!-- This should link to a Dataset Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. --> |
|
|
|
[berkeley-nest/Nectar] |
|
|
|
|
|
## Evaluation |
|
|
|
MMLU Pro 0-shot: 0.2927 |
|
|
|
|
|
#### Evaluation Data |
|
|
|
<!-- This should link to a Dataset Card if possible. --> |
|
|
|
[TIGER-AI-Lab/MMLU-Pro] |
|
|
|
|
|
## Environmental Impact |
|
|
|
<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly --> |
|
|
|
Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700). |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
