Text Generation
Transformers
TensorBoard
Safetensors
gemma2
Generated from Trainer
text-generation-inference
Instructions to use JuIm/ProGemma with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use JuIm/ProGemma with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="JuIm/ProGemma")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("JuIm/ProGemma") model = AutoModelForCausalLM.from_pretrained("JuIm/ProGemma") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use JuIm/ProGemma with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "JuIm/ProGemma" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "JuIm/ProGemma", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/JuIm/ProGemma
- SGLang
How to use JuIm/ProGemma 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 "JuIm/ProGemma" \ --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": "JuIm/ProGemma", "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 "JuIm/ProGemma" \ --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": "JuIm/ProGemma", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use JuIm/ProGemma with Docker Model Runner:
docker model run hf.co/JuIm/ProGemma
Update config.json
Browse files- config.json +3 -3
config.json
CHANGED
|
@@ -6,9 +6,9 @@
|
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
"attn_logit_softcapping": 50.0,
|
| 9 |
-
"bos_token_id":
|
| 10 |
"cache_implementation": "hybrid",
|
| 11 |
-
"eos_token_id":
|
| 12 |
"final_logit_softcapping": 30.0,
|
| 13 |
"head_dim": 256,
|
| 14 |
"hidden_activation": "gelu_pytorch_tanh",
|
|
@@ -20,7 +20,7 @@
|
|
| 20 |
"num_attention_heads": 16,
|
| 21 |
"num_hidden_layers": 14,
|
| 22 |
"num_key_value_heads": 16,
|
| 23 |
-
"pad_token_id":
|
| 24 |
"query_pre_attn_scalar": 224,
|
| 25 |
"rms_norm_eps": 1e-06,
|
| 26 |
"rope_theta": 10000.0,
|
|
|
|
| 6 |
"attention_bias": false,
|
| 7 |
"attention_dropout": 0.0,
|
| 8 |
"attn_logit_softcapping": 50.0,
|
| 9 |
+
"bos_token_id": 20,
|
| 10 |
"cache_implementation": "hybrid",
|
| 11 |
+
"eos_token_id": 21,
|
| 12 |
"final_logit_softcapping": 30.0,
|
| 13 |
"head_dim": 256,
|
| 14 |
"hidden_activation": "gelu_pytorch_tanh",
|
|
|
|
| 20 |
"num_attention_heads": 16,
|
| 21 |
"num_hidden_layers": 14,
|
| 22 |
"num_key_value_heads": 16,
|
| 23 |
+
"pad_token_id": 22,
|
| 24 |
"query_pre_attn_scalar": 224,
|
| 25 |
"rms_norm_eps": 1e-06,
|
| 26 |
"rope_theta": 10000.0,
|