Instructions to use ishanjmukherjee/llama2-1m-pg19 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ishanjmukherjee/llama2-1m-pg19 with Transformers:

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

pipe = pipeline("text-generation", model="ishanjmukherjee/llama2-1m-pg19")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ishanjmukherjee/llama2-1m-pg19")
model = AutoModelForCausalLM.from_pretrained("ishanjmukherjee/llama2-1m-pg19", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use ishanjmukherjee/llama2-1m-pg19 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ishanjmukherjee/llama2-1m-pg19"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ishanjmukherjee/llama2-1m-pg19",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ishanjmukherjee/llama2-1m-pg19

SGLang

How to use ishanjmukherjee/llama2-1m-pg19 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 "ishanjmukherjee/llama2-1m-pg19" \
    --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": "ishanjmukherjee/llama2-1m-pg19",
		"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 "ishanjmukherjee/llama2-1m-pg19" \
        --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": "ishanjmukherjee/llama2-1m-pg19",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ishanjmukherjee/llama2-1m-pg19 with Docker Model Runner:
```
docker model run hf.co/ishanjmukherjee/llama2-1m-pg19
```

llama2-1m-pg19

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 4.0628

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: 0.001
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: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
lr_scheduler_warmup_steps: 500
num_epochs: 1

Training results

Training Loss	Epoch	Step	Validation Loss
5.5529	0.0412	500	5.5682
4.4825	0.0825	1000	4.5591
4.2184	0.1237	1500	4.3646
4.2034	0.1650	2000	4.2814
4.0955	0.2062	2500	4.2316
4.1191	0.2475	3000	4.2016
4.0088	0.2887	3500	4.1736
4.0641	0.3299	4000	4.1580
4.1002	0.3712	4500	4.1433
4.0197	0.4124	5000	4.1292
3.9741	0.4537	5500	4.1164
3.9915	0.4949	6000	4.1134
4.01	0.5361	6500	4.1027
3.9424	0.5774	7000	4.0973
4.0078	0.6186	7500	4.0894
4.0254	0.6599	8000	4.0856
3.9711	0.7011	8500	4.0815
3.9905	0.7424	9000	4.0774
3.9657	0.7836	9500	4.0743
3.9494	0.8248	10000	4.0699
4.0339	0.8661	10500	4.0691
3.9739	0.9073	11000	4.0659
3.9678	0.9486	11500	4.0643
3.9043	0.9898	12000	4.0628

Framework versions

Transformers 4.51.3
Pytorch 2.7.1+cu126
Datasets 4.0.0
Tokenizers 0.21.2

Downloads last month: 5

Safetensors

Model size

2.11M params

Tensor type

F32