Instructions to use stabilityai/tiny-random-stablelm-2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use stabilityai/tiny-random-stablelm-2 with Transformers:

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

pipe = pipeline("text-generation", model="stabilityai/tiny-random-stablelm-2")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("stabilityai/tiny-random-stablelm-2")
model = AutoModelForCausalLM.from_pretrained("stabilityai/tiny-random-stablelm-2")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use stabilityai/tiny-random-stablelm-2 with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/stabilityai/tiny-random-stablelm-2

SGLang

How to use stabilityai/tiny-random-stablelm-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 "stabilityai/tiny-random-stablelm-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": "stabilityai/tiny-random-stablelm-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 "stabilityai/tiny-random-stablelm-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": "stabilityai/tiny-random-stablelm-2",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use stabilityai/tiny-random-stablelm-2 with Docker Model Runner:
```
docker model run hf.co/stabilityai/tiny-random-stablelm-2
```

jon-tow commited on Mar 31, 2024

Commit

8adff3b

verified ·

1 Parent(s): 25ddf6d

update(weights): train rand init model for 5k steps on cosmopedia

Browse files

Files changed (3) hide show

config.json +11 -8
generation_config.json +2 -2
model.safetensors +2 -2

config.json CHANGED Viewed

@@ -1,30 +1,33 @@
 {
   "architectures": [
     "StableLmForCausalLM"
   ],
   "attention_dropout": 0.0,
-  "bos_token_id": 0,
-  "eos_token_id": 0,
   "hidden_act": "silu",
   "hidden_dropout": 0.0,
   "hidden_size": 512,
-  "initializer_range": 0.02,
-  "intermediate_size": 64,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 4096,
   "model_type": "stablelm",
   "num_attention_heads": 16,
-  "num_hidden_layers": 2,
-  "num_key_value_heads": 2,
   "partial_rotary_factor": 0.25,
   "qk_layernorm": true,
   "rope_scaling": null,
   "rope_theta": 10000,
   "tie_word_embeddings": false,
-  "torch_dtype": "float32",
   "transformers_version": "4.39.0.dev0",
   "use_cache": true,
   "use_parallel_residual": true,
   "use_qkv_bias": false,
-  "vocab_size": 50304
 }

 {
+  "_name_or_path": "/weka2/home-guac/stable-lm/stable-neox/ckpts/stablelm_2_tiny/global_step6000",
   "architectures": [
     "StableLmForCausalLM"
   ],
   "attention_dropout": 0.0,
+  "bos_token_id": 100257,
+  "eos_token_id": 100257,
   "hidden_act": "silu",
   "hidden_dropout": 0.0,
   "hidden_size": 512,
+  "initializer_range": 0.01,
+  "intermediate_size": 1536,
   "layer_norm_eps": 1e-05,
   "max_position_embeddings": 4096,
   "model_type": "stablelm",
   "num_attention_heads": 16,
+  "num_hidden_layers": 8,
+  "num_key_value_heads": 4,
   "partial_rotary_factor": 0.25,
   "qk_layernorm": true,
   "rope_scaling": null,
   "rope_theta": 10000,
+  "rotary_scaling_factor": 1.0,
   "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
   "transformers_version": "4.39.0.dev0",
   "use_cache": true,
+  "use_norm_bias": false,
   "use_parallel_residual": true,
   "use_qkv_bias": false,
+  "vocab_size": 100352
 }

generation_config.json CHANGED Viewed

@@ -1,6 +1,6 @@
 {
   "_from_model_config": true,
-  "bos_token_id": 0,
-  "eos_token_id": 0,
   "transformers_version": "4.39.0.dev0"
 }

 {
   "_from_model_config": true,
+  "bos_token_id": 100257,
+  "eos_token_id": 100257,
   "transformers_version": "4.39.0.dev0"
 }

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:43482736fab1c471b5112f63e63b92a96244153a058bdb710d3617188ff90672
-size 211573792

 version https://git-lfs.github.com/spec/v1
+oid sha256:5d04033878a589c26fea21e661a7f251d493790ac6bd595ad16a5834e6b17b5f
+size 253811856