Downloads · 30 days
3
8% of all-time downloads
llama-3.1-8b-sdft-5e-6lr
cfierro/llama-3.1-8b-sdft-5e-6lr
llama-3.1-8b-sdft-5e-6lr is a machine learning model from cfierro. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
- Total steps: 500 - Best metric: None - Best checkpoint: None
All-time downloads
37
Public
Parameters
266K
112 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt96.4 GB · 86%
At a glance
- Model type
- llama
- Access
- Public
- Created
- Mar 10, 2026
- Updated
- Mar 10, 2026
- SHA
- c8cf399d
Model card
From the Hugging Face model README
llama-3.1-8b-sdft-5e-6lr
Training Hyperparameters
| Parameter | Value |
|---|---|
| learning_rate | 5e-06 |
| num_train_epochs | 1 |
| per_device_train_batch_size | 1 |
| gradient_accumulation_steps | 2 |
| weight_decay | 0.0 |
| warmup_ratio | 0.03 |
| warmup_steps | 0 |
| lr_scheduler_type | SchedulerType.CONSTANT |
| optim | OptimizerNames.ADAMW_TORCH_FUSED |
| bf16 | True |
| fp16 | False |
| max_grad_norm | 1 |
| max_steps | 500 |
| save_steps | 100 |
| deepspeed | /home/it4i-cfierro/weight-steering/deepspeed_configs/zero3_offload.json |
| gradient_checkpointing | True |
Training Results
- Total steps: 500
- Best metric: None
- Best checkpoint: None
- Type
- llama
- Created
- Mar 10, 2026
- Updated
- Mar 10, 2026