Instructions to use fxmarty/tiny-llama-fast-tokenizer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use fxmarty/tiny-llama-fast-tokenizer with Transformers:

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

pipe = pipeline("text-generation", model="fxmarty/tiny-llama-fast-tokenizer")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("fxmarty/tiny-llama-fast-tokenizer")
model = AutoModelForCausalLM.from_pretrained("fxmarty/tiny-llama-fast-tokenizer")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use fxmarty/tiny-llama-fast-tokenizer with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/fxmarty/tiny-llama-fast-tokenizer

SGLang

How to use fxmarty/tiny-llama-fast-tokenizer 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 "fxmarty/tiny-llama-fast-tokenizer" \
    --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": "fxmarty/tiny-llama-fast-tokenizer",
		"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 "fxmarty/tiny-llama-fast-tokenizer" \
        --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": "fxmarty/tiny-llama-fast-tokenizer",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use fxmarty/tiny-llama-fast-tokenizer with Docker Model Runner:
```
docker model run hf.co/fxmarty/tiny-llama-fast-tokenizer
```

Félix Marty commited on Jun 12, 2023

Commit

c11267e

1 Parent(s): daae59b

fix tokenizer?

Browse files

Files changed (3) hide show

special_tokens_map.json +3 -3
tokenizer.json +7 -1
tokenizer_config.json +4 -6

special_tokens_map.json CHANGED Viewed

@@ -1,20 +1,20 @@
 {
   "bos_token": {
-    "content": "",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,
     "single_word": false
   },
   "eos_token": {
-    "content": "",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,
     "single_word": false
   },
   "unk_token": {
-    "content": "",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,

 {
   "bos_token": {
+    "content": "<s>",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,
     "single_word": false
   },
   "eos_token": {
+    "content": "</s>",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,
     "single_word": false
   },
   "unk_token": {
+    "content": "<unk>",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,

tokenizer.json CHANGED Viewed

@@ -77,10 +77,16 @@
           "type_id": 0
         }
       },
       {
         "Sequence": {
           "id": "B",
-          "type_id": 0
         }
       }
     ],

           "type_id": 0
         }
       },
+      {
+        "SpecialToken": {
+          "id": "<s>",
+          "type_id": 1
+        }
+      },
       {
         "Sequence": {
           "id": "B",
+          "type_id": 1
         }
       }
     ],

tokenizer_config.json CHANGED Viewed

@@ -1,9 +1,7 @@
 {
-  "add_bos_token": true,
-  "add_eos_token": false,
   "bos_token": {
     "__type": "AddedToken",
-    "content": "",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,
@@ -12,19 +10,19 @@
   "clean_up_tokenization_spaces": false,
   "eos_token": {
     "__type": "AddedToken",
-    "content": "",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,
     "single_word": false
   },
-  "model_max_length": 1000000000000000019884624838656,
   "pad_token": null,
   "sp_model_kwargs": {},
   "tokenizer_class": "LlamaTokenizer",
   "unk_token": {
     "__type": "AddedToken",
-    "content": "",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,

 {
   "bos_token": {
     "__type": "AddedToken",
+    "content": "<s>",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,
   "clean_up_tokenization_spaces": false,
   "eos_token": {
     "__type": "AddedToken",
+    "content": "</s>",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,
     "single_word": false
   },
+  "model_max_length": 2048,
   "pad_token": null,
   "sp_model_kwargs": {},
   "tokenizer_class": "LlamaTokenizer",
   "unk_token": {
     "__type": "AddedToken",
+    "content": "<unk>",
     "lstrip": false,
     "normalized": true,
     "rstrip": false,