Instructions to use PHIMemo/llama31-8b-instruct-sft-balanced-3k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use PHIMemo/llama31-8b-instruct-sft-balanced-3k with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="PHIMemo/llama31-8b-instruct-sft-balanced-3k")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("PHIMemo/llama31-8b-instruct-sft-balanced-3k", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use PHIMemo/llama31-8b-instruct-sft-balanced-3k with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "PHIMemo/llama31-8b-instruct-sft-balanced-3k"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PHIMemo/llama31-8b-instruct-sft-balanced-3k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/PHIMemo/llama31-8b-instruct-sft-balanced-3k

SGLang

How to use PHIMemo/llama31-8b-instruct-sft-balanced-3k with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "PHIMemo/llama31-8b-instruct-sft-balanced-3k" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PHIMemo/llama31-8b-instruct-sft-balanced-3k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "PHIMemo/llama31-8b-instruct-sft-balanced-3k" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "PHIMemo/llama31-8b-instruct-sft-balanced-3k",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use PHIMemo/llama31-8b-instruct-sft-balanced-3k with Docker Model Runner:
```
docker model run hf.co/PHIMemo/llama31-8b-instruct-sft-balanced-3k
```

kparvataneni commited on 4 days ago

Commit

d85ec16

verified ·

1 Parent(s): a3b9ba5

Update model card with all SFT checkpoint revisions

Browse files

Files changed (1) hide show

README.md +17 -7

README.md CHANGED Viewed

@@ -11,23 +11,33 @@ library_name: transformers
 # Llama 3.1 8B Instruct — balanced 3k SFT (synthetic canaries)
-Fine-tuned checkpoint from the **model-memo-diff** / PHIMemo project (PHI / benign / public seeded documents).
 ## Revisions
-| Revision | Training step | Role |
 |---|---|---|
-| `step-000160` | 160 | Early / low-memorization donor |
-| `step-000760` | 760 | Near-peak memorization |
 ## Load
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 repo = "PHIMemo/llama31-8b-instruct-sft-balanced-3k"
-tok = AutoTokenizer.from_pretrained(repo, revision="step-000760")
-model = AutoModelForCausalLM.from_pretrained(repo, revision="step-000760", torch_dtype="auto", device_map="auto")
 ```
 ## Notes
 - Synthetic dossiers only (not real PHI).
-- Weights only (`model.safetensors`); optimizer / trainer state not uploaded.
 - Gated base model terms still apply for Llama 3.1.

 # Llama 3.1 8B Instruct — balanced 3k SFT (synthetic canaries)
+Fine-tuned checkpoints from the **model-memo-diff** / PHIMemo project (PHI / benign / public seeded documents).
 ## Revisions
+| Revision | Training step | Notes |
 |---|---|---|
+| `step-000040` | 40 | SFT checkpoint |
+| `step-000080` | 80 | SFT checkpoint |
+| `step-000160` | 160 | early / donor |
+| `step-000240` | 240 | SFT checkpoint |
+| `step-000320` | 320 | SFT checkpoint |
+| `step-000400` | 400 | SFT checkpoint |
+| `step-000480` | 480 | SFT checkpoint |
+| `step-000640` | 640 | SFT checkpoint |
+| `step-000720` | 720 | SFT checkpoint |
+| `step-000760` | 760 | near-peak |
+| `step-000800` | 800 | near-peak |
 ## Load
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 repo = "PHIMemo/llama31-8b-instruct-sft-balanced-3k"
+rev = "step-000800"  # or any revision above
+tok = AutoTokenizer.from_pretrained(repo, revision=rev)
+model = AutoModelForCausalLM.from_pretrained(repo, revision=rev, torch_dtype="auto", device_map="auto")
 ```
 ## Notes
 - Synthetic dossiers only (not real PHI).
+- Weights only (`model.safetensors` + configs); optimizer / trainer state not uploaded.
 - Gated base model terms still apply for Llama 3.1.