Instructions to use helloadhavan/llara1.0-100M-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use helloadhavan/llara1.0-100M-base with Transformers:

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

pipe = pipeline("text-generation", model="helloadhavan/llara1.0-100M-base")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("helloadhavan/llara1.0-100M-base")
model = AutoModelForCausalLM.from_pretrained("helloadhavan/llara1.0-100M-base")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use helloadhavan/llara1.0-100M-base with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "helloadhavan/llara1.0-100M-base"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "helloadhavan/llara1.0-100M-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/helloadhavan/llara1.0-100M-base

SGLang

How to use helloadhavan/llara1.0-100M-base 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 "helloadhavan/llara1.0-100M-base" \
    --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": "helloadhavan/llara1.0-100M-base",
		"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 "helloadhavan/llara1.0-100M-base" \
        --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": "helloadhavan/llara1.0-100M-base",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use helloadhavan/llara1.0-100M-base with Docker Model Runner:
```
docker model run hf.co/helloadhavan/llara1.0-100M-base
```

helloadhavan commited on 15 days ago

Commit

23d5376

verified ·

1 Parent(s): c91529d

Update README.md

Browse files

Files changed (1) hide show

README.md +5 -2

README.md CHANGED Viewed

@@ -8,6 +8,7 @@ tags:
 - text-generation
 - from-scratch
 - fineweb
 library_name: transformers
 pipeline_tag: text-generation
 ---
@@ -22,6 +23,8 @@ but it does use GPT's tokenizer
 The name **Llara** is original and unrelated to LLaMA or LoRA.
 ---
 ## Model Details
@@ -35,8 +38,8 @@ The name **Llara** is original and unrelated to LLaMA or LoRA.
 | Layers | 12 |
 | Attention heads | 12 |
 | Vocabulary | 50,257 (GPT-2 BPE) |
-| Training data | FineWeb (HuggingFaceFW/fineweb), Custom dataset |
-| Training docs | 1,000,000 documents |
 | Epochs | 1 |
 | Precision | fp16 |

 - text-generation
 - from-scratch
 - fineweb
+- undertrained
 library_name: transformers
 pipeline_tag: text-generation
 ---
 The name **Llara** is original and unrelated to LLaMA or LoRA.
+**Note**: The model is undertrained according to `The Chinchilla Laws (2022)`
 ---
 ## Model Details
 | Layers | 12 |
 | Attention heads | 12 |
 | Vocabulary | 50,257 (GPT-2 BPE) |
+| Training data | FineWeb (HuggingFaceFW/fineweb) + Custom dataset |
+| Training docs | 256,000,000 tokens |
 | Epochs | 1 |
 | Precision | fp16 |