Instructions to use codys12/BitJamba-init with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use codys12/BitJamba-init with Transformers:

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

pipe = pipeline("text-generation", model="codys12/BitJamba-init", trust_remote_code=True)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("codys12/BitJamba-init", trust_remote_code=True)
model = AutoModelForCausalLM.from_pretrained("codys12/BitJamba-init", trust_remote_code=True)

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use codys12/BitJamba-init with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "codys12/BitJamba-init"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "codys12/BitJamba-init",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/codys12/BitJamba-init

SGLang

How to use codys12/BitJamba-init 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 "codys12/BitJamba-init" \
    --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": "codys12/BitJamba-init",
		"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 "codys12/BitJamba-init" \
        --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": "codys12/BitJamba-init",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use codys12/BitJamba-init with Docker Model Runner:
```
docker model run hf.co/codys12/BitJamba-init
```

BitJamba-init / config.json

codys12

Upload JambaForCausalLM

0a70213 verified almost 2 years ago

raw

history blame contribute delete

1.4 kB

	{
	"_name_or_path": "ai21labs/Jamba-v0.1",
	"architectures": [
	"JambaForCausalLM"
	],
	"attention_dropout": 0.0,
	"attn_layer_offset": 4,
	"attn_layer_period": 8,
	"auto_map": {
	"AutoConfig": "ai21labs/Jamba-v0.1--configuration_jamba.JambaConfig",
	"AutoModel": "ai21labs/Jamba-v0.1--modeling_jamba.JambaModel",
	"AutoModelForCausalLM": "ai21labs/Jamba-v0.1--modeling_jamba.JambaForCausalLM",
	"AutoModelForSequenceClassification": "ai21labs/Jamba-v0.1--model.JambaForSequenceClassification"
	},
	"bos_token_id": 1,
	"eos_token_id": 2,
	"expert_layer_offset": 1,
	"expert_layer_period": 2,
	"hidden_act": "silu",
	"hidden_size": 4096,
	"initializer_range": 0.02,
	"intermediate_size": 14336,
	"mamba_conv_bias": true,
	"mamba_d_conv": 4,
	"mamba_d_state": 16,
	"mamba_dt_rank": 256,
	"mamba_expand": 2,
	"mamba_proj_bias": false,
	"max_position_embeddings": 262144,
	"model_type": "jamba",
	"num_attention_heads": 32,
	"num_experts": 16,
	"num_experts_per_tok": 2,
	"num_hidden_layers": 32,
	"num_key_value_heads": 8,
	"num_logits_to_keep": 1,
	"output_router_logits": false,
	"pad_token_id": 0,
	"rms_norm_eps": 1e-06,
	"router_aux_loss_coef": 0.001,
	"sliding_window": null,
	"tie_word_embeddings": false,
	"torch_dtype": "float32",
	"transformers_version": "4.41.0",
	"use_cache": true,
	"use_mamba_kernels": true,
	"vocab_size": 65536
	}