Downloads · 30 days
12
14% of all-time downloads
llama3.1_8b_sft-solo-attn-v2-k28
doupari/llama3.1_8b_sft-solo-attn-v2-k28
llama3.1_8b_sft-solo-attn-v2-k28 is a machine learning model from doupari. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
This folder was prepared from the DeepSpeed ZeRO checkpoint at /home/hyeseojeon/SparseLayer-LLM/Capsule/Capsule/modelfrombaseofficial/llama3.18bsft-solo-attn-v2-k28/epoch1.
All-time downloads
87
Public
Repo size
64.3 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin32.1 GB · 100%
At a glance
- Model type
- llama
- Access
- Public
- Created
- Apr 30, 2026
- Updated
- May 1, 2026
- SHA
- e11c249f
Model card
From the Hugging Face model README
llama3.1_8b_sft-solo-attn-v2-k28
This folder was prepared from the DeepSpeed ZeRO checkpoint at /home/hyeseojeon/SparseLayer-LLM/Capsule/Capsule/model_from_base_official/llama3.1_8b_sft-solo-attn-v2-k28/epoch_1.
Load
from transformers import AutoModelForCausalLM, AutoTokenizer
model_id = "/home/hyeseojeon/SparseLayer-LLM/Capsule/Capsule/model_from_base_official/llama3.1_8b_sft-solo-attn-v2-k28/final"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype="auto")
Notes
backbone.jsonpoints tometa-llama/Llama-3.1-8B.- Tokenizer and chat template files were copied from
/home/hyeseojeon/SparseLayer-LLM/Capsule/Capsule/model_from_base_official/llama3.1_8b_sft-solo-attn-v2-k24/final. - To upload only the Hugging Face package files, use:
python /home/hyeseojeon/SparseLayer-LLM/Capsule/upload_hf_package.py \
--local-dir "/home/hyeseojeon/SparseLayer-LLM/Capsule/Capsule/model_from_base_official/llama3.1_8b_sft-solo-attn-v2-k28/final" \
--repo-id your-name/your-repo \
--commit-message "Upload llama3.1_8b_sft-solo-attn-v2-k28"
- Type
- llama
- Created
- Apr 30, 2026
- Updated
- May 1, 2026