Instructions to use EnmingZhang/PSALM_stage1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use EnmingZhang/PSALM_stage1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="EnmingZhang/PSALM_stage1")# Load model directly from transformers import AutoModelForCausalLM model = AutoModelForCausalLM.from_pretrained("EnmingZhang/PSALM_stage1", device_map="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use EnmingZhang/PSALM_stage1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "EnmingZhang/PSALM_stage1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "EnmingZhang/PSALM_stage1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/EnmingZhang/PSALM_stage1
- SGLang
How to use EnmingZhang/PSALM_stage1 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 "EnmingZhang/PSALM_stage1" \ --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": "EnmingZhang/PSALM_stage1", "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 "EnmingZhang/PSALM_stage1" \ --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": "EnmingZhang/PSALM_stage1", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use EnmingZhang/PSALM_stage1 with Docker Model Runner:
docker model run hf.co/EnmingZhang/PSALM_stage1
init model
Browse files- config.json +44 -0
- mm_projector.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,44 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/mnt/output/pretrained_models/phi-1_5_dev",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"PhiForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 1,
|
| 8 |
+
"embd_pdrop": 0.0,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"freeze_mm_mlp_adapter": false,
|
| 11 |
+
"hidden_act": "gelu_new",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"image_aspect_ratio": "square",
|
| 14 |
+
"image_grid_pinpoints": null,
|
| 15 |
+
"initializer_range": 0.02,
|
| 16 |
+
"intermediate_size": 8192,
|
| 17 |
+
"layer_norm_eps": 1e-05,
|
| 18 |
+
"max_position_embeddings": 2048,
|
| 19 |
+
"mm_hidden_size": 256,
|
| 20 |
+
"mm_input_embeds": 1024,
|
| 21 |
+
"mm_projector_type": "swin_conv",
|
| 22 |
+
"mm_use_im_patch_token": false,
|
| 23 |
+
"mm_use_im_start_end": false,
|
| 24 |
+
"mm_vision_tower": "/path/to/swinbase/model.pkl",
|
| 25 |
+
"model_type": "llava_phi",
|
| 26 |
+
"num_attention_heads": 32,
|
| 27 |
+
"num_hidden_layers": 24,
|
| 28 |
+
"partial_rotary_factor": 0.5,
|
| 29 |
+
"pretraining_tp": 1,
|
| 30 |
+
"projector_outdim": 2048,
|
| 31 |
+
"qk_layernorm": false,
|
| 32 |
+
"resid_pdrop": 0.0,
|
| 33 |
+
"rope_scaling": null,
|
| 34 |
+
"rope_theta": 10000.0,
|
| 35 |
+
"swin_type": "base",
|
| 36 |
+
"tie_word_embeddings": false,
|
| 37 |
+
"transformers_version": "4.36.2",
|
| 38 |
+
"tune_mm_mlp_adapter": true,
|
| 39 |
+
"use_cache": true,
|
| 40 |
+
"use_mm_proj": true,
|
| 41 |
+
"vocab_size": 50296,
|
| 42 |
+
"with_layernorm": false,
|
| 43 |
+
"with_norm": true
|
| 44 |
+
}
|
mm_projector.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6edaacf5e1173ed1450798e1a00f3c5e3448697ec2b30dbe33b18397b2370d10
|
| 3 |
+
size 125861573
|