Instructions to use RichardErkhov/deepset_-_tinyroberta-squad2-4bits with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RichardErkhov/deepset_-_tinyroberta-squad2-4bits with Transformers:

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

pipe = pipeline("text-generation", model="RichardErkhov/deepset_-_tinyroberta-squad2-4bits")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("RichardErkhov/deepset_-_tinyroberta-squad2-4bits")
model = AutoModelForCausalLM.from_pretrained("RichardErkhov/deepset_-_tinyroberta-squad2-4bits")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use RichardErkhov/deepset_-_tinyroberta-squad2-4bits with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RichardErkhov/deepset_-_tinyroberta-squad2-4bits"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RichardErkhov/deepset_-_tinyroberta-squad2-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/RichardErkhov/deepset_-_tinyroberta-squad2-4bits

SGLang

How to use RichardErkhov/deepset_-_tinyroberta-squad2-4bits 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 "RichardErkhov/deepset_-_tinyroberta-squad2-4bits" \
    --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": "RichardErkhov/deepset_-_tinyroberta-squad2-4bits",
		"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 "RichardErkhov/deepset_-_tinyroberta-squad2-4bits" \
        --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": "RichardErkhov/deepset_-_tinyroberta-squad2-4bits",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use RichardErkhov/deepset_-_tinyroberta-squad2-4bits with Docker Model Runner:
```
docker model run hf.co/RichardErkhov/deepset_-_tinyroberta-squad2-4bits
```

RichardErkhov commited on May 2, 2024

Commit

8b972da

verified ·

1 Parent(s): b60c27c

uploaded model

Browse files

Files changed (1) hide show

config.json +45 -0

config.json ADDED Viewed

	@@ -0,0 +1,45 @@

+{
+  "_name_or_path": "tinyroberta-squad2",
+  "architectures": [
+    "RobertaForCausalLM"
+  ],
+  "attention_probs_dropout_prob": 0.1,
+  "bos_token_id": 0,
+  "classifier_dropout": null,
+  "eos_token_id": 2,
+  "gradient_checkpointing": false,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.1,
+  "hidden_size": 768,
+  "initializer_range": 0.02,
+  "intermediate_size": 3072,
+  "language": "english",
+  "layer_norm_eps": 1e-05,
+  "max_position_embeddings": 514,
+  "model_type": "roberta",
+  "name": "Roberta",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 6,
+  "pad_token_id": 1,
+  "position_embedding_type": "absolute",
+  "quantization_config": {
+    "_load_in_4bit": true,
+    "_load_in_8bit": false,
+    "bnb_4bit_compute_dtype": "float32",
+    "bnb_4bit_quant_storage": "uint8",
+    "bnb_4bit_quant_type": "fp4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": true,
+    "load_in_8bit": false,
+    "quant_method": "bitsandbytes"
+  },
+  "torch_dtype": "float16",
+  "transformers_version": "4.39.3",
+  "type_vocab_size": 1,
+  "use_cache": true,
+  "vocab_size": 50265
+}