Instructions to use ninagroot/Llama-360M-RUN1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use ninagroot/Llama-360M-RUN1 with Transformers:

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

pipe = pipeline("text-generation", model="ninagroot/Llama-360M-RUN1")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("ninagroot/Llama-360M-RUN1")
model = AutoModelForCausalLM.from_pretrained("ninagroot/Llama-360M-RUN1")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use ninagroot/Llama-360M-RUN1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "ninagroot/Llama-360M-RUN1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "ninagroot/Llama-360M-RUN1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/ninagroot/Llama-360M-RUN1

SGLang

How to use ninagroot/Llama-360M-RUN1 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 "ninagroot/Llama-360M-RUN1" \
    --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": "ninagroot/Llama-360M-RUN1",
		"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 "ninagroot/Llama-360M-RUN1" \
        --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": "ninagroot/Llama-360M-RUN1",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use ninagroot/Llama-360M-RUN1 with Docker Model Runner:
```
docker model run hf.co/ninagroot/Llama-360M-RUN1
```

ninagroot commited on Apr 2, 2024

Commit

ce55a4b

verified ·

1 Parent(s): fb54902

ninagroot/Llama-360Mtest

Browse files

Files changed (4) hide show

README.md +11 -11
model.safetensors +1 -1
runs/Apr02_15-25-19_gcn25.local.snellius.surf.nl/events.out.tfevents.1712064331.gcn25.local.snellius.surf.nl.270804.0 +3 -0
training_args.bin +1 -1

README.md CHANGED Viewed

@@ -13,7 +13,7 @@ should probably proofread and complete it, then remove this comment. -->
 This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.5562
 ## Model description
@@ -33,11 +33,11 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 0.0003
-- train_batch_size: 4
 - eval_batch_size: 8
 - seed: 42
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 8
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 100
@@ -48,14 +48,14 @@ The following hyperparameters were used during training:
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 4.0358        | 1.0   | 238  | 4.2148          |
-| 2.9242        | 2.0   | 476  | 3.8319          |
-| 2.1979        | 3.0   | 714  | 3.7139          |
-| 1.6277        | 4.0   | 952  | 3.6317          |
-| 1.1634        | 5.0   | 1190 | 3.5924          |
-| 0.6908        | 6.0   | 1428 | 3.5634          |
-| 0.5032        | 7.0   | 1666 | 3.5571          |
-| 0.4221        | 8.0   | 1904 | 3.5562          |
 ### Framework versions

 This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
 It achieves the following results on the evaluation set:
+- Loss: 3.4045
 ## Model description
 The following hyperparameters were used during training:
 - learning_rate: 0.0003
+- train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
 - gradient_accumulation_steps: 2
+- total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_steps: 100
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
+| 7.4576        | 0.99  | 59   | 5.4054          |
+| 3.927         | 2.0   | 119  | 3.9632          |
+| 2.8202        | 2.99  | 178  | 3.6557          |
+| 1.9108        | 4.0   | 238  | 3.4939          |
+| 1.3277        | 4.99  | 297  | 3.4565          |
+| 0.8717        | 6.0   | 357  | 3.4181          |
+| 0.5598        | 6.99  | 416  | 3.4027          |
+| 0.4604        | 7.93  | 472  | 3.4045          |
 ### Framework versions

model.safetensors CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1de2b1182a90a2acd55aab0446cab00886d578f14b90bba5d823581b03182bd3
 size 1570992472

 version https://git-lfs.github.com/spec/v1
+oid sha256:b419164dc271226dca8b2eb7ddfda6791f97b0dad4f38b7505bd781adfee7dd9
 size 1570992472

runs/Apr02_15-25-19_gcn25.local.snellius.surf.nl/events.out.tfevents.1712064331.gcn25.local.snellius.surf.nl.270804.0 ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:3db561dc0e414d0061119f0b2e847ccaf209dac54b60f46710914c091c350b0a
+size 11904

training_args.bin CHANGED Viewed

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9e7b7191f1016592d40aedefde600b24e11b911e88bad036a0e569fc4f5a206b
 size 4984

 version https://git-lfs.github.com/spec/v1
+oid sha256:2dff12963512f842bbd1b0d3dd9a75580e2b1d112c698330839007a214f43af6
 size 4984