Instructions to use normalcomputing/extended-mind-llama-2-7b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use normalcomputing/extended-mind-llama-2-7b with Transformers:

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

pipe = pipeline("text-generation", model="normalcomputing/extended-mind-llama-2-7b", trust_remote_code=True)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("normalcomputing/extended-mind-llama-2-7b", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use normalcomputing/extended-mind-llama-2-7b with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "normalcomputing/extended-mind-llama-2-7b"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "normalcomputing/extended-mind-llama-2-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/normalcomputing/extended-mind-llama-2-7b

SGLang

How to use normalcomputing/extended-mind-llama-2-7b 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 "normalcomputing/extended-mind-llama-2-7b" \
    --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": "normalcomputing/extended-mind-llama-2-7b",
		"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 "normalcomputing/extended-mind-llama-2-7b" \
        --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": "normalcomputing/extended-mind-llama-2-7b",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use normalcomputing/extended-mind-llama-2-7b with Docker Model Runner:
```
docker model run hf.co/normalcomputing/extended-mind-llama-2-7b
```

phoebeklett commited on May 30, 2024

Commit

0860325

verified ·

1 Parent(s): 0354e72

Upload 2 files

Browse files

Files changed (1) hide show

modeling.py +4 -4

modeling.py CHANGED Viewed

@@ -47,7 +47,7 @@ from transformers.utils import (
     replace_return_docstrings,
 )
-from .configuration import ExtendedLlamaConfig
 logger = logging.get_logger(__name__)
@@ -1144,7 +1144,7 @@ class ExtendedLlamaForCausalLM(LlamaPreTrainedModel):
     _tied_weights_keys = ["lm_head.weight"]
-    def __init__(self, config, external_memories=None):
         super().__init__(config)
         self.model = ExtendedLlamaModel(config)
         self.vocab_size = config.vocab_size
@@ -1242,9 +1242,9 @@ class ExtendedLlamaForCausalLM(LlamaPreTrainedModel):
         if (
             self.memory_ids is not None and self.memories is None
         ):
             self.memories = self.generate_cache(
-                torch.tensor(self.memory_ids, device=self.device),
-                cache_type=self.memory_type,
             )
             # EM: Remove special tokens from memory cache
             if self.remove_special_ids:

     replace_return_docstrings,
 )
+from emts_clean.src.llama.configuration import ExtendedLlamaConfig
 logger = logging.get_logger(__name__)
     _tied_weights_keys = ["lm_head.weight"]
+    def __init__(self, config, external_memories:list=None):
         super().__init__(config)
         self.model = ExtendedLlamaModel(config)
         self.vocab_size = config.vocab_size
         if (
             self.memory_ids is not None and self.memories is None
         ):
+            self.memory_ids = torch.tensor([self.memory_ids], device=self.device) if type(self.memory_ids)==list else self.memory_ids
             self.memories = self.generate_cache(
+                self.memory_ids, cache_type=self.memory_type,
             )
             # EM: Remove special tokens from memory cache
             if self.remove_special_ids: