Downloads · 30 days
0
Meta-Llama-3-8B-Instruct-2024-07-21-15-01
CLASS-MATE/Meta-Llama-3-8B-Instruct-2024-07-21-15-01
Meta-Llama-3-8B-Instruct-2024-07-21-15-01 is a machine learning model from CLASS-MATE. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
outputdir="./finetuning-results", overwriteoutputdir=True, pushtohub=True, pushtohubmodelid=basemodel.split("/")[-1] + "-" + datetime.datetime.now().strftime("%Y-%m-%d-%H-%M"), pushtohuborganization="CLASS-MATE", push…
Access
Public
Updated Jul 22, 2024
Repo size
109 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors109 MB · 92%
At a glance
- Access
- Public
- Created
- Jul 21, 2024
- Updated
- Jul 22, 2024
- SHA
- 04393d4d
Model card
From the Hugging Face model README
output_dir="./finetuning-results", overwrite_output_dir=True, push_to_hub=True, push_to_hub_model_id=base_model.split("/")[-1] + "-" + datetime.datetime.now().strftime("%Y-%m-%d-%H-%M"), push_to_hub_organization="CLASS-MATE", push_to_hub_token="TOKEN", num_train_epochs=1, per_device_train_batch_size=8, gradient_accumulation_steps=1, optim="paged_adamw_32bit", # save_steps=500, save_strategy="epoch", logging_steps=50, evaluation_strategy="steps", eval_steps=500, learning_rate=0.1, weight_decay=0.001, fp16=False, bf16=False, max_grad_norm=0.3, # max_steps=-1, warmup_ratio=0.03, group_by_length=True, lr_scheduler_type="constant",
- Created
- Jul 21, 2024
- Updated
- Jul 22, 2024