Instructions to use bjoernp/llama-2-7b-de-instruct_v0.2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use bjoernp/llama-2-7b-de-instruct_v0.2 with Transformers:

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

pipe = pipeline("text-generation", model="bjoernp/llama-2-7b-de-instruct_v0.2", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("bjoernp/llama-2-7b-de-instruct_v0.2", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("bjoernp/llama-2-7b-de-instruct_v0.2", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use bjoernp/llama-2-7b-de-instruct_v0.2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bjoernp/llama-2-7b-de-instruct_v0.2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "bjoernp/llama-2-7b-de-instruct_v0.2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/bjoernp/llama-2-7b-de-instruct_v0.2

SGLang

How to use bjoernp/llama-2-7b-de-instruct_v0.2 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 "bjoernp/llama-2-7b-de-instruct_v0.2" \
    --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": "bjoernp/llama-2-7b-de-instruct_v0.2",
		"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 "bjoernp/llama-2-7b-de-instruct_v0.2" \
        --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": "bjoernp/llama-2-7b-de-instruct_v0.2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use bjoernp/llama-2-7b-de-instruct_v0.2 with Docker Model Runner:
```
docker model run hf.co/bjoernp/llama-2-7b-de-instruct_v0.2
```

bjoernp commited on Aug 23, 2023

Commit

5d8cbb5

1 Parent(s): 2a62b40

Training in progress, epoch 1

Browse files

Files changed (2) hide show

config.json +32 -0
training_args.bin +3 -0

config.json ADDED Viewed

	@@ -0,0 +1,32 @@

+{
+  "_name_or_path": "bjoernp/llama-2-7b-de",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "auto_map": {
+    "AutoModelForCausalLM": "bjoernp/llama-2-7b-de--modeling_flash_llama.LlamaForCausalLM"
+  },
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 8192,
+  "model_type": "llama",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": {
+    "factor": 2.0,
+    "type": "linear"
+  },
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.33.0.dev0",
+  "use_cache": false,
+  "vocab_size": 32000
+}

training_args.bin ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:7ec7ceb1d650f6db832218daa72de6593e7c2f782d9b7793350813c030d5afdd
+size 4155