Instructions to use mnoukhov/pythia1b-sft-tldr with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mnoukhov/pythia1b-sft-tldr with Transformers:

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

pipe = pipeline("text-generation", model="mnoukhov/pythia1b-sft-tldr")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mnoukhov/pythia1b-sft-tldr")
model = AutoModelForCausalLM.from_pretrained("mnoukhov/pythia1b-sft-tldr")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use mnoukhov/pythia1b-sft-tldr with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mnoukhov/pythia1b-sft-tldr"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mnoukhov/pythia1b-sft-tldr",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mnoukhov/pythia1b-sft-tldr

SGLang

How to use mnoukhov/pythia1b-sft-tldr 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 "mnoukhov/pythia1b-sft-tldr" \
    --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": "mnoukhov/pythia1b-sft-tldr",
		"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 "mnoukhov/pythia1b-sft-tldr" \
        --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": "mnoukhov/pythia1b-sft-tldr",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mnoukhov/pythia1b-sft-tldr with Docker Model Runner:
```
docker model run hf.co/mnoukhov/pythia1b-sft-tldr
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

pythia1b-sft-tldr

This model is a fine-tuned version of EleutherAI/pythia-1b-deduped on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 2.3828

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 1e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
distributed_type: multi-GPU
num_devices: 4
gradient_accumulation_steps: 4
total_train_batch_size: 128
total_eval_batch_size: 32
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: cosine
num_epochs: 1
mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss
2.5278	0.2007	183	2.4199
2.4136	0.4013	366	2.4004
2.3978	0.6020	549	2.3887
2.3813	0.8026	732	2.3828

Framework versions

Transformers 4.41.1
Pytorch 2.2.1+cu121
Datasets 2.19.0
Tokenizers 0.19.1

Downloads last month: 5

Safetensors

Model size

1B params

Tensor type

F16

Model tree for mnoukhov/pythia1b-sft-tldr

Base model

EleutherAI/pythia-1b-deduped

Finetuned

(16)

this model

Finetunes

1 model

Quantizations

1 model

Collection including mnoukhov/pythia1b-sft-tldr

Asynchronous RLHF

Collection

Models and datasets for asynchronous rlhf paper, see code at https://github.com/mnoukhov/async_rlhf • 10 items • Updated Oct 28, 2024