Instructions to use Gustav-Proxi/SalienceFormer-Gemma2B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Gustav-Proxi/SalienceFormer-Gemma2B with Transformers:

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

pipe = pipeline("text-generation", model="Gustav-Proxi/SalienceFormer-Gemma2B")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("Gustav-Proxi/SalienceFormer-Gemma2B", dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Gustav-Proxi/SalienceFormer-Gemma2B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Gustav-Proxi/SalienceFormer-Gemma2B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Gustav-Proxi/SalienceFormer-Gemma2B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Gustav-Proxi/SalienceFormer-Gemma2B

SGLang

How to use Gustav-Proxi/SalienceFormer-Gemma2B 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 "Gustav-Proxi/SalienceFormer-Gemma2B" \
    --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": "Gustav-Proxi/SalienceFormer-Gemma2B",
		"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 "Gustav-Proxi/SalienceFormer-Gemma2B" \
        --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": "Gustav-Proxi/SalienceFormer-Gemma2B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Gustav-Proxi/SalienceFormer-Gemma2B with Docker Model Runner:
```
docker model run hf.co/Gustav-Proxi/SalienceFormer-Gemma2B
```

SalienceFormer-Gemma2B

SalienceFormer is a biologically-inspired memory architecture that brings hippocampal memory consolidation to large language models. This model integrates hippocampal mechanisms directly into the Gemma-2B transformer.

Model Description

SalienceFormer adds three key components inspired by how the human hippocampus processes memories:

Component	Inspiration	Function
Salience Gate	Sharp Wave Ripples (SPW-Rs)	Dual-pathway importance scoring
Memory Buffer	Sleep Replay	Priority-based consolidation
Drift Calibrator	Synaptic Homeostasis	Embedding stability

Architecture

Input Tokens -> Gemma-2B (frozen + LoRA) -> Hidden States
    -> Salience Gate (importance scoring)
    -> Drift Calibrator (stability)
    -> Memory Buffer (consolidation)
    -> Output Fusion (cross-attention)
    -> Output Logits

Results

Perplexity (WikiText-2)

Model	Parameters	Perplexity
GPT-2	124M	29.41
Gemma-2B	2B	~18
SalienceFormer	2B + 15M	11.83

Ablation Study

Configuration	PPL	Impact
Full SalienceFormer	11.83	baseline
No Salience Gate	39.75	+27.92
No Memory Buffer	89.84	+78.01

Brain-Like Behavior

Metric	Value	Interpretation
Content/Function Ratio	2.11x	Selective memory (content words tagged more)
Long-Range Benefit	+6.95 PPL	Better context retention
Buffer Priority	4.9/5.0	High-importance retention

Usage

from salienceformer import SalienceFormer, SalienceFormerConfig
from huggingface_hub import hf_hub_download
import torch

# Download checkpoint
ckpt_path = hf_hub_download(
    repo_id="Gustav-Proxi/SalienceFormer-Gemma2B",
    filename="pytorch_model.pt"
)

# Initialize model
config = SalienceFormerConfig(
    base_model_name="google/gemma-2b",
    freeze_base=True,
    use_lora=True,
)
model = SalienceFormer(config)

# Load weights
ckpt = torch.load(ckpt_path, map_location="cpu")
model.load_state_dict(ckpt["model_state_dict"], strict=False)

# Generate
from transformers import AutoTokenizer
tokenizer = AutoTokenizer.from_pretrained("google/gemma-2b")
inputs = tokenizer("The capital of France is", return_tensors="pt")
outputs = model.generate(inputs["input_ids"], max_new_tokens=20)
print(tokenizer.decode(outputs[0]))

Training

Base Model: Gemma-2B (frozen with LoRA)
Dataset: WikiText-2
Hardware: NVIDIA RTX 4090 (24GB)
Training Time: ~24 hours
Best Checkpoint: step-110000

Citation

@misc{salienceformer2025,
  title={SalienceFormer: Salience-Gated Memory Consolidation for Transformers},
  author={Vaishak Girish Kumar and Sanika},
  year={2025},
  howpublished={\url{https://github.com/Gustav-Proxi/SalienceFormer}},
}

License

Apache 2.0

Downloads last month: -; Downloads are not tracked for this model. How to track

Model tree for Gustav-Proxi/SalienceFormer-Gemma2B

Base model

google/gemma-2b

Finetuned

(300)

this model

Gustav-Proxi
/

SalienceFormer-Gemma2B