Instructions to use baffo32/genji-python-6B-split with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use baffo32/genji-python-6B-split with Transformers:

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

pipe = pipeline("text-generation", model="baffo32/genji-python-6B-split")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("baffo32/genji-python-6B-split")
model = AutoModelForCausalLM.from_pretrained("baffo32/genji-python-6B-split")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use baffo32/genji-python-6B-split with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "baffo32/genji-python-6B-split"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "baffo32/genji-python-6B-split",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/baffo32/genji-python-6B-split

SGLang

How to use baffo32/genji-python-6B-split 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 "baffo32/genji-python-6B-split" \
    --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": "baffo32/genji-python-6B-split",
		"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 "baffo32/genji-python-6B-split" \
        --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": "baffo32/genji-python-6B-split",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use baffo32/genji-python-6B-split with Docker Model Runner:
```
docker model run hf.co/baffo32/genji-python-6B-split
```

baffo32 commited on Aug 21, 2021

Commit

85b14e2

1 Parent(s): 890390b

add alternative config + keys is_split, subfolder

Browse files

Files changed (2) hide show

config.json +51 -0
model/config.json +2 -1

config.json ADDED Viewed

	@@ -0,0 +1,51 @@

+{
+    "activation_function": "gelu_new",
+    "architectures": [
+      "GPTNeoForCausalLM"
+    ],
+    "attention_dropout": 0,
+    "attention_layers": ["global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global"],
+    "attention_types": [
+      [
+        [
+          "global"
+        ],
+        28
+      ]
+    ],
+    "bos_token_id": 50256,
+    "embed_dropout": 0,
+    "eos_token_id": 50256,
+    "gradient_checkpointing": false,
+    "hidden_size": 4096,
+    "initializer_range": 0.02,
+    "intermediate_size": null,
+    "is_split": true,
+    "jax": true,
+    "layer_norm_epsilon": 1e-05,
+    "max_position_embeddings": 2048,
+    "model_type": "gpt_neo",
+    "num_heads": 16,
+    "num_layers": 28,
+    "resid_dropout": 0,
+    "rotary": true,
+    "rotary_dim": 64,
+    "subfolder": "model",
+    "summary_activation": null,
+    "summary_first_dropout": 0.1,
+    "summary_proj_to_labels": true,
+    "summary_type": "cls_index",
+    "summary_use_proj": true,
+    "transformers_version": "4.10.0.dev0",
+    "use_cache": true,
+    "vocab_size": 50400,
+    "window_size": 256,
+    "tokenizer_class": "GPT2Tokenizer",
+    "task_specific_params": {
+      "text-generation": {
+        "do_sample": true,
+        "temperature": 1.0,
+        "max_length": 50
+      }
+    }
+  }

model/config.json CHANGED Viewed

@@ -20,6 +20,7 @@
     "hidden_size": 4096,
     "initializer_range": 0.02,
     "intermediate_size": null,
     "jax": true,
     "layer_norm_epsilon": 1e-05,
     "max_position_embeddings": 2048,
@@ -46,4 +47,4 @@
         "max_length": 50
       }
     }
-  }

     "hidden_size": 4096,
     "initializer_range": 0.02,
     "intermediate_size": null,
+    "is_split": true,
     "jax": true,
     "layer_norm_epsilon": 1e-05,
     "max_position_embeddings": 2048,
         "max_length": 50
       }
     }
+  }