Instructions to use Technotech/RedPajama-Base-3B-4bit-128g with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Technotech/RedPajama-Base-3B-4bit-128g with Transformers:

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

pipe = pipeline("text-generation", model="Technotech/RedPajama-Base-3B-4bit-128g")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("Technotech/RedPajama-Base-3B-4bit-128g")
model = AutoModelForCausalLM.from_pretrained("Technotech/RedPajama-Base-3B-4bit-128g")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use Technotech/RedPajama-Base-3B-4bit-128g with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Technotech/RedPajama-Base-3B-4bit-128g"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Technotech/RedPajama-Base-3B-4bit-128g",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Technotech/RedPajama-Base-3B-4bit-128g

SGLang

How to use Technotech/RedPajama-Base-3B-4bit-128g 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 "Technotech/RedPajama-Base-3B-4bit-128g" \
    --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": "Technotech/RedPajama-Base-3B-4bit-128g",
		"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 "Technotech/RedPajama-Base-3B-4bit-128g" \
        --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": "Technotech/RedPajama-Base-3B-4bit-128g",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Technotech/RedPajama-Base-3B-4bit-128g with Docker Model Runner:
```
docker model run hf.co/Technotech/RedPajama-Base-3B-4bit-128g
```

Technotech commited on Jun 12, 2023

Commit

bf7aa2b

1 Parent(s): d50d127

AutoGPTQ model for togethercomputer/RedPajama-INCITE-Base-3B-v1: 4bits, gr128, desc_act=False

Browse files

Files changed (3) hide show

config.json +26 -0
gptq_model-4bit-128g.safetensors +3 -0
quantize_config.json +10 -0

config.json ADDED Viewed

	@@ -0,0 +1,26 @@

+{
+  "_name_or_path": "togethercomputer/RedPajama-INCITE-Base-3B-v1",
+  "architectures": [
+    "GPTNeoXForCausalLM"
+  ],
+  "bos_token_id": 0,
+  "classifier_dropout": 0.1,
+  "eos_token_id": 0,
+  "hidden_act": "gelu",
+  "hidden_size": 2560,
+  "initializer_range": 0.02,
+  "intermediate_size": 10240,
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 2048,
+  "model_type": "gpt_neox",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "rotary_emb_base": 10000,
+  "rotary_pct": 1.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.29.2",
+  "use_cache": true,
+  "use_parallel_residual": false,
+  "vocab_size": 50432
+}

gptq_model-4bit-128g.safetensors ADDED Viewed

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

+version https://git-lfs.github.com/spec/v1
+oid sha256:680d50004d5525c8134f20165007aa21fd3c65843891c202afd288c349bb1064
+size 1962625112

quantize_config.json ADDED Viewed

	@@ -0,0 +1,10 @@

+{
+  "bits": 4,
+  "group_size": 128,
+  "damp_percent": 0.01,
+  "desc_act": false,
+  "sym": true,
+  "true_sequential": true,
+  "model_name_or_path": null,
+  "model_file_base_name": null
+}