Instructions to use eshmoideas/open-dev-gen with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use eshmoideas/open-dev-gen with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="eshmoideas/open-dev-gen") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("eshmoideas/open-dev-gen") model = AutoModelForCausalLM.from_pretrained("eshmoideas/open-dev-gen") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use eshmoideas/open-dev-gen with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "eshmoideas/open-dev-gen" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "eshmoideas/open-dev-gen", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/eshmoideas/open-dev-gen
- SGLang
How to use eshmoideas/open-dev-gen 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 "eshmoideas/open-dev-gen" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "eshmoideas/open-dev-gen", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'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 "eshmoideas/open-dev-gen" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "eshmoideas/open-dev-gen", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use eshmoideas/open-dev-gen with Docker Model Runner:
docker model run hf.co/eshmoideas/open-dev-gen
File size: 19,658 Bytes
8b71769 | 1 | {"embedding.weight": "BF16", "block.0.attn.norm.scale": "BF16", "block.0.attn.qkv.weight": "BF16", "block.0.attn.qkv.bias": "BF16", "block.0.attn.sinks": "BF16", "block.0.attn.out.weight": "BF16", "block.0.attn.out.bias": "BF16", "block.0.mlp.norm.scale": "BF16", "block.0.mlp.gate.weight": "BF16", "block.0.mlp.gate.bias": "BF16", "block.0.mlp.mlp1_weight.blocks": "FP4", "block.0.mlp.mlp1_weight.scales": "UE8", "block.0.mlp.mlp1_bias": "BF16", "block.0.mlp.mlp2_weight.blocks": "FP4", "block.0.mlp.mlp2_weight.scales": "UE8", "block.0.mlp.mlp2_bias": "BF16", "block.1.attn.norm.scale": "BF16", "block.1.attn.qkv.weight": "BF16", "block.1.attn.qkv.bias": "BF16", "block.1.attn.sinks": "BF16", "block.1.attn.out.weight": "BF16", "block.1.attn.out.bias": "BF16", "block.1.mlp.norm.scale": "BF16", "block.1.mlp.gate.weight": "BF16", "block.1.mlp.gate.bias": "BF16", "block.1.mlp.mlp1_weight.blocks": "FP4", "block.1.mlp.mlp1_weight.scales": "UE8", "block.1.mlp.mlp1_bias": "BF16", "block.1.mlp.mlp2_weight.blocks": "FP4", "block.1.mlp.mlp2_weight.scales": "UE8", "block.1.mlp.mlp2_bias": "BF16", "block.2.attn.norm.scale": "BF16", "block.2.attn.qkv.weight": "BF16", "block.2.attn.qkv.bias": "BF16", "block.2.attn.sinks": "BF16", "block.2.attn.out.weight": "BF16", "block.2.attn.out.bias": "BF16", "block.2.mlp.norm.scale": "BF16", "block.2.mlp.gate.weight": "BF16", "block.2.mlp.gate.bias": "BF16", "block.2.mlp.mlp1_weight.blocks": "FP4", "block.2.mlp.mlp1_weight.scales": "UE8", "block.2.mlp.mlp1_bias": "BF16", "block.2.mlp.mlp2_weight.blocks": "FP4", "block.2.mlp.mlp2_weight.scales": "UE8", "block.2.mlp.mlp2_bias": "BF16", "block.3.attn.norm.scale": "BF16", "block.3.attn.qkv.weight": "BF16", "block.3.attn.qkv.bias": "BF16", "block.3.attn.sinks": "BF16", "block.3.attn.out.weight": "BF16", "block.3.attn.out.bias": "BF16", "block.3.mlp.norm.scale": "BF16", "block.3.mlp.gate.weight": "BF16", "block.3.mlp.gate.bias": "BF16", "block.3.mlp.mlp1_weight.blocks": "FP4", "block.3.mlp.mlp1_weight.scales": "UE8", "block.3.mlp.mlp1_bias": "BF16", "block.3.mlp.mlp2_weight.blocks": "FP4", "block.3.mlp.mlp2_weight.scales": "UE8", "block.3.mlp.mlp2_bias": "BF16", "block.4.attn.norm.scale": "BF16", "block.4.attn.qkv.weight": "BF16", "block.4.attn.qkv.bias": "BF16", "block.4.attn.sinks": "BF16", "block.4.attn.out.weight": "BF16", "block.4.attn.out.bias": "BF16", "block.4.mlp.norm.scale": "BF16", "block.4.mlp.gate.weight": "BF16", "block.4.mlp.gate.bias": "BF16", "block.4.mlp.mlp1_weight.blocks": "FP4", "block.4.mlp.mlp1_weight.scales": "UE8", "block.4.mlp.mlp1_bias": "BF16", "block.4.mlp.mlp2_weight.blocks": "FP4", "block.4.mlp.mlp2_weight.scales": "UE8", "block.4.mlp.mlp2_bias": "BF16", "block.5.attn.norm.scale": "BF16", "block.5.attn.qkv.weight": "BF16", "block.5.attn.qkv.bias": "BF16", "block.5.attn.sinks": "BF16", "block.5.attn.out.weight": "BF16", "block.5.attn.out.bias": "BF16", "block.5.mlp.norm.scale": "BF16", "block.5.mlp.gate.weight": "BF16", "block.5.mlp.gate.bias": "BF16", "block.5.mlp.mlp1_weight.blocks": "FP4", "block.5.mlp.mlp1_weight.scales": "UE8", "block.5.mlp.mlp1_bias": "BF16", "block.5.mlp.mlp2_weight.blocks": "FP4", "block.5.mlp.mlp2_weight.scales": "UE8", "block.5.mlp.mlp2_bias": "BF16", "block.6.attn.norm.scale": "BF16", "block.6.attn.qkv.weight": "BF16", "block.6.attn.qkv.bias": "BF16", "block.6.attn.sinks": "BF16", "block.6.attn.out.weight": "BF16", "block.6.attn.out.bias": "BF16", "block.6.mlp.norm.scale": "BF16", "block.6.mlp.gate.weight": "BF16", "block.6.mlp.gate.bias": "BF16", "block.6.mlp.mlp1_weight.blocks": "FP4", "block.6.mlp.mlp1_weight.scales": "UE8", "block.6.mlp.mlp1_bias": "BF16", "block.6.mlp.mlp2_weight.blocks": "FP4", "block.6.mlp.mlp2_weight.scales": "UE8", "block.6.mlp.mlp2_bias": "BF16", "block.7.attn.norm.scale": "BF16", "block.7.attn.qkv.weight": "BF16", "block.7.attn.qkv.bias": "BF16", "block.7.attn.sinks": "BF16", "block.7.attn.out.weight": "BF16", "block.7.attn.out.bias": "BF16", "block.7.mlp.norm.scale": "BF16", "block.7.mlp.gate.weight": "BF16", "block.7.mlp.gate.bias": "BF16", "block.7.mlp.mlp1_weight.blocks": "FP4", "block.7.mlp.mlp1_weight.scales": "UE8", "block.7.mlp.mlp1_bias": "BF16", "block.7.mlp.mlp2_weight.blocks": "FP4", "block.7.mlp.mlp2_weight.scales": "UE8", "block.7.mlp.mlp2_bias": "BF16", "block.8.attn.norm.scale": "BF16", "block.8.attn.qkv.weight": "BF16", "block.8.attn.qkv.bias": "BF16", "block.8.attn.sinks": "BF16", "block.8.attn.out.weight": "BF16", "block.8.attn.out.bias": "BF16", "block.8.mlp.norm.scale": "BF16", "block.8.mlp.gate.weight": "BF16", "block.8.mlp.gate.bias": "BF16", "block.8.mlp.mlp1_weight.blocks": "FP4", "block.8.mlp.mlp1_weight.scales": "UE8", "block.8.mlp.mlp1_bias": "BF16", "block.8.mlp.mlp2_weight.blocks": "FP4", "block.8.mlp.mlp2_weight.scales": "UE8", "block.8.mlp.mlp2_bias": "BF16", "block.9.attn.norm.scale": "BF16", "block.9.attn.qkv.weight": "BF16", "block.9.attn.qkv.bias": "BF16", "block.9.attn.sinks": "BF16", "block.9.attn.out.weight": "BF16", "block.9.attn.out.bias": "BF16", "block.9.mlp.norm.scale": "BF16", "block.9.mlp.gate.weight": "BF16", "block.9.mlp.gate.bias": "BF16", "block.9.mlp.mlp1_weight.blocks": "FP4", "block.9.mlp.mlp1_weight.scales": "UE8", "block.9.mlp.mlp1_bias": "BF16", "block.9.mlp.mlp2_weight.blocks": "FP4", "block.9.mlp.mlp2_weight.scales": "UE8", "block.9.mlp.mlp2_bias": "BF16", "block.10.attn.norm.scale": "BF16", "block.10.attn.qkv.weight": "BF16", "block.10.attn.qkv.bias": "BF16", "block.10.attn.sinks": "BF16", "block.10.attn.out.weight": "BF16", "block.10.attn.out.bias": "BF16", "block.10.mlp.norm.scale": "BF16", "block.10.mlp.gate.weight": "BF16", "block.10.mlp.gate.bias": "BF16", "block.10.mlp.mlp1_weight.blocks": "FP4", "block.10.mlp.mlp1_weight.scales": "UE8", "block.10.mlp.mlp1_bias": "BF16", "block.10.mlp.mlp2_weight.blocks": "FP4", "block.10.mlp.mlp2_weight.scales": "UE8", "block.10.mlp.mlp2_bias": "BF16", "block.11.attn.norm.scale": "BF16", "block.11.attn.qkv.weight": "BF16", "block.11.attn.qkv.bias": "BF16", "block.11.attn.sinks": "BF16", "block.11.attn.out.weight": "BF16", "block.11.attn.out.bias": "BF16", "block.11.mlp.norm.scale": "BF16", "block.11.mlp.gate.weight": "BF16", "block.11.mlp.gate.bias": "BF16", "block.11.mlp.mlp1_weight.blocks": "FP4", "block.11.mlp.mlp1_weight.scales": "UE8", "block.11.mlp.mlp1_bias": "BF16", "block.11.mlp.mlp2_weight.blocks": "FP4", "block.11.mlp.mlp2_weight.scales": "UE8", "block.11.mlp.mlp2_bias": "BF16", "block.12.attn.norm.scale": "BF16", "block.12.attn.qkv.weight": "BF16", "block.12.attn.qkv.bias": "BF16", "block.12.attn.sinks": "BF16", "block.12.attn.out.weight": "BF16", "block.12.attn.out.bias": "BF16", "block.12.mlp.norm.scale": "BF16", "block.12.mlp.gate.weight": "BF16", "block.12.mlp.gate.bias": "BF16", "block.12.mlp.mlp1_weight.blocks": "FP4", "block.12.mlp.mlp1_weight.scales": "UE8", "block.12.mlp.mlp1_bias": "BF16", "block.12.mlp.mlp2_weight.blocks": "FP4", "block.12.mlp.mlp2_weight.scales": "UE8", "block.12.mlp.mlp2_bias": "BF16", "block.13.attn.norm.scale": "BF16", "block.13.attn.qkv.weight": "BF16", "block.13.attn.qkv.bias": "BF16", "block.13.attn.sinks": "BF16", "block.13.attn.out.weight": "BF16", "block.13.attn.out.bias": "BF16", "block.13.mlp.norm.scale": "BF16", "block.13.mlp.gate.weight": "BF16", "block.13.mlp.gate.bias": "BF16", "block.13.mlp.mlp1_weight.blocks": "FP4", "block.13.mlp.mlp1_weight.scales": "UE8", "block.13.mlp.mlp1_bias": "BF16", "block.13.mlp.mlp2_weight.blocks": "FP4", "block.13.mlp.mlp2_weight.scales": "UE8", "block.13.mlp.mlp2_bias": "BF16", "block.14.attn.norm.scale": "BF16", "block.14.attn.qkv.weight": "BF16", "block.14.attn.qkv.bias": "BF16", "block.14.attn.sinks": "BF16", "block.14.attn.out.weight": "BF16", "block.14.attn.out.bias": "BF16", "block.14.mlp.norm.scale": "BF16", "block.14.mlp.gate.weight": "BF16", "block.14.mlp.gate.bias": "BF16", "block.14.mlp.mlp1_weight.blocks": "FP4", "block.14.mlp.mlp1_weight.scales": "UE8", "block.14.mlp.mlp1_bias": "BF16", "block.14.mlp.mlp2_weight.blocks": "FP4", "block.14.mlp.mlp2_weight.scales": "UE8", "block.14.mlp.mlp2_bias": "BF16", "block.15.attn.norm.scale": "BF16", "block.15.attn.qkv.weight": "BF16", "block.15.attn.qkv.bias": "BF16", "block.15.attn.sinks": "BF16", "block.15.attn.out.weight": "BF16", "block.15.attn.out.bias": "BF16", "block.15.mlp.norm.scale": "BF16", "block.15.mlp.gate.weight": "BF16", "block.15.mlp.gate.bias": "BF16", "block.15.mlp.mlp1_weight.blocks": "FP4", "block.15.mlp.mlp1_weight.scales": "UE8", "block.15.mlp.mlp1_bias": "BF16", "block.15.mlp.mlp2_weight.blocks": "FP4", "block.15.mlp.mlp2_weight.scales": "UE8", "block.15.mlp.mlp2_bias": "BF16", "block.16.attn.norm.scale": "BF16", "block.16.attn.qkv.weight": "BF16", "block.16.attn.qkv.bias": "BF16", "block.16.attn.sinks": "BF16", "block.16.attn.out.weight": "BF16", "block.16.attn.out.bias": "BF16", "block.16.mlp.norm.scale": "BF16", "block.16.mlp.gate.weight": "BF16", "block.16.mlp.gate.bias": "BF16", "block.16.mlp.mlp1_weight.blocks": "FP4", "block.16.mlp.mlp1_weight.scales": "UE8", "block.16.mlp.mlp1_bias": "BF16", "block.16.mlp.mlp2_weight.blocks": "FP4", "block.16.mlp.mlp2_weight.scales": "UE8", "block.16.mlp.mlp2_bias": "BF16", "block.17.attn.norm.scale": "BF16", "block.17.attn.qkv.weight": "BF16", "block.17.attn.qkv.bias": "BF16", "block.17.attn.sinks": "BF16", "block.17.attn.out.weight": "BF16", "block.17.attn.out.bias": "BF16", "block.17.mlp.norm.scale": "BF16", "block.17.mlp.gate.weight": "BF16", "block.17.mlp.gate.bias": "BF16", "block.17.mlp.mlp1_weight.blocks": "FP4", "block.17.mlp.mlp1_weight.scales": "UE8", "block.17.mlp.mlp1_bias": "BF16", "block.17.mlp.mlp2_weight.blocks": "FP4", "block.17.mlp.mlp2_weight.scales": "UE8", "block.17.mlp.mlp2_bias": "BF16", "block.18.attn.norm.scale": "BF16", "block.18.attn.qkv.weight": "BF16", "block.18.attn.qkv.bias": "BF16", "block.18.attn.sinks": "BF16", "block.18.attn.out.weight": "BF16", "block.18.attn.out.bias": "BF16", "block.18.mlp.norm.scale": "BF16", "block.18.mlp.gate.weight": "BF16", "block.18.mlp.gate.bias": "BF16", "block.18.mlp.mlp1_weight.blocks": "FP4", "block.18.mlp.mlp1_weight.scales": "UE8", "block.18.mlp.mlp1_bias": "BF16", "block.18.mlp.mlp2_weight.blocks": "FP4", "block.18.mlp.mlp2_weight.scales": "UE8", "block.18.mlp.mlp2_bias": "BF16", "block.19.attn.norm.scale": "BF16", "block.19.attn.qkv.weight": "BF16", "block.19.attn.qkv.bias": "BF16", "block.19.attn.sinks": "BF16", "block.19.attn.out.weight": "BF16", "block.19.attn.out.bias": "BF16", "block.19.mlp.norm.scale": "BF16", "block.19.mlp.gate.weight": "BF16", "block.19.mlp.gate.bias": "BF16", "block.19.mlp.mlp1_weight.blocks": "FP4", "block.19.mlp.mlp1_weight.scales": "UE8", "block.19.mlp.mlp1_bias": "BF16", "block.19.mlp.mlp2_weight.blocks": "FP4", "block.19.mlp.mlp2_weight.scales": "UE8", "block.19.mlp.mlp2_bias": "BF16", "block.20.attn.norm.scale": "BF16", "block.20.attn.qkv.weight": "BF16", "block.20.attn.qkv.bias": "BF16", "block.20.attn.sinks": "BF16", "block.20.attn.out.weight": "BF16", "block.20.attn.out.bias": "BF16", "block.20.mlp.norm.scale": "BF16", "block.20.mlp.gate.weight": "BF16", "block.20.mlp.gate.bias": "BF16", "block.20.mlp.mlp1_weight.blocks": "FP4", "block.20.mlp.mlp1_weight.scales": "UE8", "block.20.mlp.mlp1_bias": "BF16", "block.20.mlp.mlp2_weight.blocks": "FP4", "block.20.mlp.mlp2_weight.scales": "UE8", "block.20.mlp.mlp2_bias": "BF16", "block.21.attn.norm.scale": "BF16", "block.21.attn.qkv.weight": "BF16", "block.21.attn.qkv.bias": "BF16", "block.21.attn.sinks": "BF16", "block.21.attn.out.weight": "BF16", "block.21.attn.out.bias": "BF16", "block.21.mlp.norm.scale": "BF16", "block.21.mlp.gate.weight": "BF16", "block.21.mlp.gate.bias": "BF16", "block.21.mlp.mlp1_weight.blocks": "FP4", "block.21.mlp.mlp1_weight.scales": "UE8", "block.21.mlp.mlp1_bias": "BF16", "block.21.mlp.mlp2_weight.blocks": "FP4", "block.21.mlp.mlp2_weight.scales": "UE8", "block.21.mlp.mlp2_bias": "BF16", "block.22.attn.norm.scale": "BF16", "block.22.attn.qkv.weight": "BF16", "block.22.attn.qkv.bias": "BF16", "block.22.attn.sinks": "BF16", "block.22.attn.out.weight": "BF16", "block.22.attn.out.bias": "BF16", "block.22.mlp.norm.scale": "BF16", "block.22.mlp.gate.weight": "BF16", "block.22.mlp.gate.bias": "BF16", "block.22.mlp.mlp1_weight.blocks": "FP4", "block.22.mlp.mlp1_weight.scales": "UE8", "block.22.mlp.mlp1_bias": "BF16", "block.22.mlp.mlp2_weight.blocks": "FP4", "block.22.mlp.mlp2_weight.scales": "UE8", "block.22.mlp.mlp2_bias": "BF16", "block.23.attn.norm.scale": "BF16", "block.23.attn.qkv.weight": "BF16", "block.23.attn.qkv.bias": "BF16", "block.23.attn.sinks": "BF16", "block.23.attn.out.weight": "BF16", "block.23.attn.out.bias": "BF16", "block.23.mlp.norm.scale": "BF16", "block.23.mlp.gate.weight": "BF16", "block.23.mlp.gate.bias": "BF16", "block.23.mlp.mlp1_weight.blocks": "FP4", "block.23.mlp.mlp1_weight.scales": "UE8", "block.23.mlp.mlp1_bias": "BF16", "block.23.mlp.mlp2_weight.blocks": "FP4", "block.23.mlp.mlp2_weight.scales": "UE8", "block.23.mlp.mlp2_bias": "BF16", "block.24.attn.norm.scale": "BF16", "block.24.attn.qkv.weight": "BF16", "block.24.attn.qkv.bias": "BF16", "block.24.attn.sinks": "BF16", "block.24.attn.out.weight": "BF16", "block.24.attn.out.bias": "BF16", "block.24.mlp.norm.scale": "BF16", "block.24.mlp.gate.weight": "BF16", "block.24.mlp.gate.bias": "BF16", "block.24.mlp.mlp1_weight.blocks": "FP4", "block.24.mlp.mlp1_weight.scales": "UE8", "block.24.mlp.mlp1_bias": "BF16", "block.24.mlp.mlp2_weight.blocks": "FP4", "block.24.mlp.mlp2_weight.scales": "UE8", "block.24.mlp.mlp2_bias": "BF16", "block.25.attn.norm.scale": "BF16", "block.25.attn.qkv.weight": "BF16", "block.25.attn.qkv.bias": "BF16", "block.25.attn.sinks": "BF16", "block.25.attn.out.weight": "BF16", "block.25.attn.out.bias": "BF16", "block.25.mlp.norm.scale": "BF16", "block.25.mlp.gate.weight": "BF16", "block.25.mlp.gate.bias": "BF16", "block.25.mlp.mlp1_weight.blocks": "FP4", "block.25.mlp.mlp1_weight.scales": "UE8", "block.25.mlp.mlp1_bias": "BF16", "block.25.mlp.mlp2_weight.blocks": "FP4", "block.25.mlp.mlp2_weight.scales": "UE8", "block.25.mlp.mlp2_bias": "BF16", "block.26.attn.norm.scale": "BF16", "block.26.attn.qkv.weight": "BF16", "block.26.attn.qkv.bias": "BF16", "block.26.attn.sinks": "BF16", "block.26.attn.out.weight": "BF16", "block.26.attn.out.bias": "BF16", "block.26.mlp.norm.scale": "BF16", "block.26.mlp.gate.weight": "BF16", "block.26.mlp.gate.bias": "BF16", "block.26.mlp.mlp1_weight.blocks": "FP4", "block.26.mlp.mlp1_weight.scales": "UE8", "block.26.mlp.mlp1_bias": "BF16", "block.26.mlp.mlp2_weight.blocks": "FP4", "block.26.mlp.mlp2_weight.scales": "UE8", "block.26.mlp.mlp2_bias": "BF16", "block.27.attn.norm.scale": "BF16", "block.27.attn.qkv.weight": "BF16", "block.27.attn.qkv.bias": "BF16", "block.27.attn.sinks": "BF16", "block.27.attn.out.weight": "BF16", "block.27.attn.out.bias": "BF16", "block.27.mlp.norm.scale": "BF16", "block.27.mlp.gate.weight": "BF16", "block.27.mlp.gate.bias": "BF16", "block.27.mlp.mlp1_weight.blocks": "FP4", "block.27.mlp.mlp1_weight.scales": "UE8", "block.27.mlp.mlp1_bias": "BF16", "block.27.mlp.mlp2_weight.blocks": "FP4", "block.27.mlp.mlp2_weight.scales": "UE8", "block.27.mlp.mlp2_bias": "BF16", "block.28.attn.norm.scale": "BF16", "block.28.attn.qkv.weight": "BF16", "block.28.attn.qkv.bias": "BF16", "block.28.attn.sinks": "BF16", "block.28.attn.out.weight": "BF16", "block.28.attn.out.bias": "BF16", "block.28.mlp.norm.scale": "BF16", "block.28.mlp.gate.weight": "BF16", "block.28.mlp.gate.bias": "BF16", "block.28.mlp.mlp1_weight.blocks": "FP4", "block.28.mlp.mlp1_weight.scales": "UE8", "block.28.mlp.mlp1_bias": "BF16", "block.28.mlp.mlp2_weight.blocks": "FP4", "block.28.mlp.mlp2_weight.scales": "UE8", "block.28.mlp.mlp2_bias": "BF16", "block.29.attn.norm.scale": "BF16", "block.29.attn.qkv.weight": "BF16", "block.29.attn.qkv.bias": "BF16", "block.29.attn.sinks": "BF16", "block.29.attn.out.weight": "BF16", "block.29.attn.out.bias": "BF16", "block.29.mlp.norm.scale": "BF16", "block.29.mlp.gate.weight": "BF16", "block.29.mlp.gate.bias": "BF16", "block.29.mlp.mlp1_weight.blocks": "FP4", "block.29.mlp.mlp1_weight.scales": "UE8", "block.29.mlp.mlp1_bias": "BF16", "block.29.mlp.mlp2_weight.blocks": "FP4", "block.29.mlp.mlp2_weight.scales": "UE8", "block.29.mlp.mlp2_bias": "BF16", "block.30.attn.norm.scale": "BF16", "block.30.attn.qkv.weight": "BF16", "block.30.attn.qkv.bias": "BF16", "block.30.attn.sinks": "BF16", "block.30.attn.out.weight": "BF16", "block.30.attn.out.bias": "BF16", "block.30.mlp.norm.scale": "BF16", "block.30.mlp.gate.weight": "BF16", "block.30.mlp.gate.bias": "BF16", "block.30.mlp.mlp1_weight.blocks": "FP4", "block.30.mlp.mlp1_weight.scales": "UE8", "block.30.mlp.mlp1_bias": "BF16", "block.30.mlp.mlp2_weight.blocks": "FP4", "block.30.mlp.mlp2_weight.scales": "UE8", "block.30.mlp.mlp2_bias": "BF16", "block.31.attn.norm.scale": "BF16", "block.31.attn.qkv.weight": "BF16", "block.31.attn.qkv.bias": "BF16", "block.31.attn.sinks": "BF16", "block.31.attn.out.weight": "BF16", "block.31.attn.out.bias": "BF16", "block.31.mlp.norm.scale": "BF16", "block.31.mlp.gate.weight": "BF16", "block.31.mlp.gate.bias": "BF16", "block.31.mlp.mlp1_weight.blocks": "FP4", "block.31.mlp.mlp1_weight.scales": "UE8", "block.31.mlp.mlp1_bias": "BF16", "block.31.mlp.mlp2_weight.blocks": "FP4", "block.31.mlp.mlp2_weight.scales": "UE8", "block.31.mlp.mlp2_bias": "BF16", "block.32.attn.norm.scale": "BF16", "block.32.attn.qkv.weight": "BF16", "block.32.attn.qkv.bias": "BF16", "block.32.attn.sinks": "BF16", "block.32.attn.out.weight": "BF16", "block.32.attn.out.bias": "BF16", "block.32.mlp.norm.scale": "BF16", "block.32.mlp.gate.weight": "BF16", "block.32.mlp.gate.bias": "BF16", "block.32.mlp.mlp1_weight.blocks": "FP4", "block.32.mlp.mlp1_weight.scales": "UE8", "block.32.mlp.mlp1_bias": "BF16", "block.32.mlp.mlp2_weight.blocks": "FP4", "block.32.mlp.mlp2_weight.scales": "UE8", "block.32.mlp.mlp2_bias": "BF16", "block.33.attn.norm.scale": "BF16", "block.33.attn.qkv.weight": "BF16", "block.33.attn.qkv.bias": "BF16", "block.33.attn.sinks": "BF16", "block.33.attn.out.weight": "BF16", "block.33.attn.out.bias": "BF16", "block.33.mlp.norm.scale": "BF16", "block.33.mlp.gate.weight": "BF16", "block.33.mlp.gate.bias": "BF16", "block.33.mlp.mlp1_weight.blocks": "FP4", "block.33.mlp.mlp1_weight.scales": "UE8", "block.33.mlp.mlp1_bias": "BF16", "block.33.mlp.mlp2_weight.blocks": "FP4", "block.33.mlp.mlp2_weight.scales": "UE8", "block.33.mlp.mlp2_bias": "BF16", "block.34.attn.norm.scale": "BF16", "block.34.attn.qkv.weight": "BF16", "block.34.attn.qkv.bias": "BF16", "block.34.attn.sinks": "BF16", "block.34.attn.out.weight": "BF16", "block.34.attn.out.bias": "BF16", "block.34.mlp.norm.scale": "BF16", "block.34.mlp.gate.weight": "BF16", "block.34.mlp.gate.bias": "BF16", "block.34.mlp.mlp1_weight.blocks": "FP4", "block.34.mlp.mlp1_weight.scales": "UE8", "block.34.mlp.mlp1_bias": "BF16", "block.34.mlp.mlp2_weight.blocks": "FP4", "block.34.mlp.mlp2_weight.scales": "UE8", "block.34.mlp.mlp2_bias": "BF16", "block.35.attn.norm.scale": "BF16", "block.35.attn.qkv.weight": "BF16", "block.35.attn.qkv.bias": "BF16", "block.35.attn.sinks": "BF16", "block.35.attn.out.weight": "BF16", "block.35.attn.out.bias": "BF16", "block.35.mlp.norm.scale": "BF16", "block.35.mlp.gate.weight": "BF16", "block.35.mlp.gate.bias": "BF16", "block.35.mlp.mlp1_weight.blocks": "FP4", "block.35.mlp.mlp1_weight.scales": "UE8", "block.35.mlp.mlp1_bias": "BF16", "block.35.mlp.mlp2_weight.blocks": "FP4", "block.35.mlp.mlp2_weight.scales": "UE8", "block.35.mlp.mlp2_bias": "BF16", "norm.scale": "BF16", "unembedding.weight": "BF16"} |