Instructions to use facebook/opt-66b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use facebook/opt-66b with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="facebook/opt-66b")# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("facebook/opt-66b") model = AutoModelForCausalLM.from_pretrained("facebook/opt-66b") - Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- vLLM
How to use facebook/opt-66b with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "facebook/opt-66b" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "facebook/opt-66b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/facebook/opt-66b
- SGLang
How to use facebook/opt-66b 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 "facebook/opt-66b" \ --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": "facebook/opt-66b", "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 "facebook/opt-66b" \ --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": "facebook/opt-66b", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use facebook/opt-66b with Docker Model Runner:
docker model run hf.co/facebook/opt-66b
| { | |
| "metadata": { | |
| "total_size": 131439403008 | |
| }, | |
| "weight_map": { | |
| "model/decoder/embed_positions/embedding": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/embed_tokens/embedding": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/final_layer_norm/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/final_layer_norm/scale": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/fc1/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/fc1/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/fc2/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/fc2/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/final_layer_norm/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/final_layer_norm/scale": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/self_attn/k_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/self_attn/k_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/self_attn/out_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/self_attn/out_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/self_attn/q_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/self_attn/q_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/self_attn/v_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/self_attn/v_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/self_attn_layer_norm/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/0/self_attn_layer_norm/scale": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/fc1/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/fc1/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/fc2/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/fc2/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/final_layer_norm/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/final_layer_norm/scale": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/self_attn/k_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/self_attn/k_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/self_attn/out_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/self_attn/out_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/self_attn/q_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/self_attn/q_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/self_attn/v_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/self_attn/v_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/self_attn_layer_norm/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/1/self_attn_layer_norm/scale": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/fc1/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/fc1/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/fc2/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/fc2/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/final_layer_norm/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/final_layer_norm/scale": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/self_attn/k_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/self_attn/k_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/self_attn/out_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/self_attn/out_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/self_attn/q_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/self_attn/q_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/self_attn/v_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/self_attn/v_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/self_attn_layer_norm/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/10/self_attn_layer_norm/scale": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/fc1/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/fc1/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/fc2/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/fc2/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/final_layer_norm/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/final_layer_norm/scale": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/self_attn/k_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/self_attn/k_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/self_attn/out_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/self_attn/out_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/self_attn/q_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/self_attn/q_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/self_attn/v_proj/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/self_attn/v_proj/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/self_attn_layer_norm/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/11/self_attn_layer_norm/scale": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/12/fc1/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/12/fc1/kernel": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/12/fc2/bias": "flax_model-00001-of-00014.msgpack", | |
| "model/decoder/layers/12/fc2/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/final_layer_norm/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/final_layer_norm/scale": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/self_attn/k_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/self_attn/k_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/self_attn/out_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/self_attn/out_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/self_attn/q_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/self_attn/q_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/self_attn/v_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/self_attn/v_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/self_attn_layer_norm/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/12/self_attn_layer_norm/scale": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/fc1/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/fc1/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/fc2/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/fc2/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/final_layer_norm/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/final_layer_norm/scale": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/self_attn/k_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/self_attn/k_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/self_attn/out_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/self_attn/out_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/self_attn/q_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/self_attn/q_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/self_attn/v_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/self_attn/v_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/self_attn_layer_norm/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/13/self_attn_layer_norm/scale": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/fc1/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/fc1/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/fc2/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/fc2/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/final_layer_norm/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/final_layer_norm/scale": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/self_attn/k_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/self_attn/k_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/self_attn/out_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/self_attn/out_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/self_attn/q_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/self_attn/q_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/self_attn/v_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/self_attn/v_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/self_attn_layer_norm/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/14/self_attn_layer_norm/scale": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/fc1/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/fc1/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/fc2/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/fc2/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/final_layer_norm/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/final_layer_norm/scale": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/self_attn/k_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/self_attn/k_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/self_attn/out_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/self_attn/out_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/self_attn/q_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/self_attn/q_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/self_attn/v_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/self_attn/v_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/self_attn_layer_norm/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/15/self_attn_layer_norm/scale": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/fc1/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/fc1/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/fc2/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/fc2/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/final_layer_norm/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/final_layer_norm/scale": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/self_attn/k_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/self_attn/k_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/self_attn/out_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/self_attn/out_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/self_attn/q_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/self_attn/q_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/self_attn/v_proj/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/self_attn/v_proj/kernel": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/self_attn_layer_norm/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/16/self_attn_layer_norm/scale": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/17/fc1/bias": "flax_model-00002-of-00014.msgpack", | |
| "model/decoder/layers/17/fc1/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/fc2/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/fc2/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/final_layer_norm/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/final_layer_norm/scale": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/self_attn/k_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/self_attn/k_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/self_attn/out_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/self_attn/out_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/self_attn/q_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/self_attn/q_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/self_attn/v_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/self_attn/v_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/self_attn_layer_norm/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/17/self_attn_layer_norm/scale": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/fc1/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/fc1/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/fc2/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/fc2/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/final_layer_norm/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/final_layer_norm/scale": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/self_attn/k_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/self_attn/k_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/self_attn/out_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/self_attn/out_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/self_attn/q_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/self_attn/q_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/self_attn/v_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/self_attn/v_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/self_attn_layer_norm/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/18/self_attn_layer_norm/scale": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/fc1/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/fc1/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/fc2/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/fc2/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/final_layer_norm/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/final_layer_norm/scale": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/self_attn/k_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/self_attn/k_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/self_attn/out_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/self_attn/out_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/self_attn/q_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/self_attn/q_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/self_attn/v_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/self_attn/v_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/self_attn_layer_norm/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/19/self_attn_layer_norm/scale": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/fc1/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/fc1/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/fc2/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/fc2/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/final_layer_norm/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/final_layer_norm/scale": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/self_attn/k_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/self_attn/k_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/self_attn/out_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/self_attn/out_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/self_attn/q_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/self_attn/q_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/self_attn/v_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/self_attn/v_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/self_attn_layer_norm/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/2/self_attn_layer_norm/scale": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/fc1/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/fc1/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/fc2/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/fc2/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/final_layer_norm/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/final_layer_norm/scale": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/self_attn/k_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/self_attn/k_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/self_attn/out_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/self_attn/out_proj/kernel": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/self_attn/q_proj/bias": "flax_model-00003-of-00014.msgpack", | |
| "model/decoder/layers/20/self_attn/q_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/20/self_attn/v_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/20/self_attn/v_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/20/self_attn_layer_norm/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/20/self_attn_layer_norm/scale": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/fc1/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/fc1/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/fc2/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/fc2/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/final_layer_norm/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/final_layer_norm/scale": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/self_attn/k_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/self_attn/k_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/self_attn/out_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/self_attn/out_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/self_attn/q_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/self_attn/q_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/self_attn/v_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/self_attn/v_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/self_attn_layer_norm/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/21/self_attn_layer_norm/scale": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/fc1/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/fc1/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/fc2/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/fc2/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/final_layer_norm/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/final_layer_norm/scale": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/self_attn/k_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/self_attn/k_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/self_attn/out_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/self_attn/out_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/self_attn/q_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/self_attn/q_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/self_attn/v_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/self_attn/v_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/self_attn_layer_norm/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/22/self_attn_layer_norm/scale": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/fc1/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/fc1/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/fc2/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/fc2/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/final_layer_norm/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/final_layer_norm/scale": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/self_attn/k_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/self_attn/k_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/self_attn/out_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/self_attn/out_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/self_attn/q_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/self_attn/q_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/self_attn/v_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/self_attn/v_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/self_attn_layer_norm/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/23/self_attn_layer_norm/scale": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/fc1/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/fc1/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/fc2/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/fc2/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/final_layer_norm/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/final_layer_norm/scale": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/self_attn/k_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/self_attn/k_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/self_attn/out_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/self_attn/out_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/self_attn/q_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/self_attn/q_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/self_attn/v_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/self_attn/v_proj/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/self_attn_layer_norm/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/24/self_attn_layer_norm/scale": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/25/fc1/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/25/fc1/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/25/fc2/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/25/fc2/kernel": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/25/final_layer_norm/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/25/final_layer_norm/scale": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/25/self_attn/k_proj/bias": "flax_model-00004-of-00014.msgpack", | |
| "model/decoder/layers/25/self_attn/k_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/25/self_attn/out_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/25/self_attn/out_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/25/self_attn/q_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/25/self_attn/q_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/25/self_attn/v_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/25/self_attn/v_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/25/self_attn_layer_norm/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/25/self_attn_layer_norm/scale": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/fc1/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/fc1/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/fc2/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/fc2/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/final_layer_norm/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/final_layer_norm/scale": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/self_attn/k_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/self_attn/k_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/self_attn/out_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/self_attn/out_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/self_attn/q_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/self_attn/q_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/self_attn/v_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/self_attn/v_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/self_attn_layer_norm/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/26/self_attn_layer_norm/scale": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/fc1/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/fc1/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/fc2/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/fc2/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/final_layer_norm/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/final_layer_norm/scale": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/self_attn/k_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/self_attn/k_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/self_attn/out_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/self_attn/out_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/self_attn/q_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/self_attn/q_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/self_attn/v_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/self_attn/v_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/self_attn_layer_norm/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/27/self_attn_layer_norm/scale": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/fc1/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/fc1/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/fc2/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/fc2/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/final_layer_norm/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/final_layer_norm/scale": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/self_attn/k_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/self_attn/k_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/self_attn/out_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/self_attn/out_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/self_attn/q_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/self_attn/q_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/self_attn/v_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/self_attn/v_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/self_attn_layer_norm/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/28/self_attn_layer_norm/scale": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/fc1/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/fc1/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/fc2/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/fc2/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/final_layer_norm/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/final_layer_norm/scale": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/self_attn/k_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/self_attn/k_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/self_attn/out_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/self_attn/out_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/self_attn/q_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/self_attn/q_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/self_attn/v_proj/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/self_attn/v_proj/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/self_attn_layer_norm/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/29/self_attn_layer_norm/scale": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/3/fc1/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/3/fc1/kernel": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/3/fc2/bias": "flax_model-00005-of-00014.msgpack", | |
| "model/decoder/layers/3/fc2/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/final_layer_norm/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/final_layer_norm/scale": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/self_attn/k_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/self_attn/k_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/self_attn/out_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/self_attn/out_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/self_attn/q_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/self_attn/q_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/self_attn/v_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/self_attn/v_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/self_attn_layer_norm/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/3/self_attn_layer_norm/scale": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/fc1/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/fc1/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/fc2/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/fc2/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/final_layer_norm/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/final_layer_norm/scale": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/self_attn/k_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/self_attn/k_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/self_attn/out_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/self_attn/out_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/self_attn/q_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/self_attn/q_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/self_attn/v_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/self_attn/v_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/self_attn_layer_norm/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/30/self_attn_layer_norm/scale": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/fc1/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/fc1/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/fc2/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/fc2/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/final_layer_norm/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/final_layer_norm/scale": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/self_attn/k_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/self_attn/k_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/self_attn/out_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/self_attn/out_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/self_attn/q_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/self_attn/q_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/self_attn/v_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/self_attn/v_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/self_attn_layer_norm/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/31/self_attn_layer_norm/scale": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/fc1/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/fc1/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/fc2/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/fc2/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/final_layer_norm/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/final_layer_norm/scale": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/self_attn/k_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/self_attn/k_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/self_attn/out_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/self_attn/out_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/self_attn/q_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/self_attn/q_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/self_attn/v_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/self_attn/v_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/self_attn_layer_norm/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/32/self_attn_layer_norm/scale": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/fc1/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/fc1/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/fc2/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/fc2/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/final_layer_norm/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/final_layer_norm/scale": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/self_attn/k_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/self_attn/k_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/self_attn/out_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/self_attn/out_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/self_attn/q_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/self_attn/q_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/self_attn/v_proj/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/self_attn/v_proj/kernel": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/self_attn_layer_norm/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/33/self_attn_layer_norm/scale": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/34/fc1/bias": "flax_model-00006-of-00014.msgpack", | |
| "model/decoder/layers/34/fc1/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/fc2/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/fc2/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/final_layer_norm/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/final_layer_norm/scale": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/self_attn/k_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/self_attn/k_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/self_attn/out_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/self_attn/out_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/self_attn/q_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/self_attn/q_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/self_attn/v_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/self_attn/v_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/self_attn_layer_norm/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/34/self_attn_layer_norm/scale": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/fc1/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/fc1/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/fc2/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/fc2/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/final_layer_norm/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/final_layer_norm/scale": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/self_attn/k_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/self_attn/k_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/self_attn/out_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/self_attn/out_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/self_attn/q_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/self_attn/q_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/self_attn/v_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/self_attn/v_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/self_attn_layer_norm/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/35/self_attn_layer_norm/scale": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/fc1/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/fc1/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/fc2/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/fc2/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/final_layer_norm/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/final_layer_norm/scale": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/self_attn/k_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/self_attn/k_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/self_attn/out_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/self_attn/out_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/self_attn/q_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/self_attn/q_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/self_attn/v_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/self_attn/v_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/self_attn_layer_norm/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/36/self_attn_layer_norm/scale": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/fc1/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/fc1/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/fc2/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/fc2/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/final_layer_norm/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/final_layer_norm/scale": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/self_attn/k_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/self_attn/k_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/self_attn/out_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/self_attn/out_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/self_attn/q_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/self_attn/q_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/self_attn/v_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/self_attn/v_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/self_attn_layer_norm/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/37/self_attn_layer_norm/scale": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/fc1/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/fc1/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/fc2/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/fc2/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/final_layer_norm/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/final_layer_norm/scale": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/self_attn/k_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/self_attn/k_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/self_attn/out_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/self_attn/out_proj/kernel": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/self_attn/q_proj/bias": "flax_model-00007-of-00014.msgpack", | |
| "model/decoder/layers/38/self_attn/q_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/38/self_attn/v_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/38/self_attn/v_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/38/self_attn_layer_norm/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/38/self_attn_layer_norm/scale": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/fc1/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/fc1/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/fc2/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/fc2/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/final_layer_norm/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/final_layer_norm/scale": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/self_attn/k_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/self_attn/k_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/self_attn/out_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/self_attn/out_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/self_attn/q_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/self_attn/q_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/self_attn/v_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/self_attn/v_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/self_attn_layer_norm/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/39/self_attn_layer_norm/scale": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/fc1/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/fc1/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/fc2/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/fc2/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/final_layer_norm/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/final_layer_norm/scale": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/self_attn/k_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/self_attn/k_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/self_attn/out_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/self_attn/out_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/self_attn/q_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/self_attn/q_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/self_attn/v_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/self_attn/v_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/self_attn_layer_norm/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/4/self_attn_layer_norm/scale": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/fc1/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/fc1/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/fc2/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/fc2/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/final_layer_norm/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/final_layer_norm/scale": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/self_attn/k_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/self_attn/k_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/self_attn/out_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/self_attn/out_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/self_attn/q_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/self_attn/q_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/self_attn/v_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/self_attn/v_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/self_attn_layer_norm/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/40/self_attn_layer_norm/scale": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/fc1/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/fc1/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/fc2/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/fc2/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/final_layer_norm/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/final_layer_norm/scale": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/self_attn/k_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/self_attn/k_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/self_attn/out_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/self_attn/out_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/self_attn/q_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/self_attn/q_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/self_attn/v_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/self_attn/v_proj/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/self_attn_layer_norm/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/41/self_attn_layer_norm/scale": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/42/fc1/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/42/fc1/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/42/fc2/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/42/fc2/kernel": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/42/final_layer_norm/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/42/final_layer_norm/scale": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/42/self_attn/k_proj/bias": "flax_model-00008-of-00014.msgpack", | |
| "model/decoder/layers/42/self_attn/k_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/42/self_attn/out_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/42/self_attn/out_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/42/self_attn/q_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/42/self_attn/q_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/42/self_attn/v_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/42/self_attn/v_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/42/self_attn_layer_norm/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/42/self_attn_layer_norm/scale": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/fc1/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/fc1/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/fc2/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/fc2/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/final_layer_norm/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/final_layer_norm/scale": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/self_attn/k_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/self_attn/k_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/self_attn/out_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/self_attn/out_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/self_attn/q_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/self_attn/q_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/self_attn/v_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/self_attn/v_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/self_attn_layer_norm/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/43/self_attn_layer_norm/scale": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/fc1/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/fc1/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/fc2/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/fc2/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/final_layer_norm/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/final_layer_norm/scale": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/self_attn/k_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/self_attn/k_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/self_attn/out_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/self_attn/out_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/self_attn/q_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/self_attn/q_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/self_attn/v_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/self_attn/v_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/self_attn_layer_norm/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/44/self_attn_layer_norm/scale": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/fc1/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/fc1/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/fc2/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/fc2/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/final_layer_norm/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/final_layer_norm/scale": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/self_attn/k_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/self_attn/k_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/self_attn/out_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/self_attn/out_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/self_attn/q_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/self_attn/q_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/self_attn/v_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/self_attn/v_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/self_attn_layer_norm/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/45/self_attn_layer_norm/scale": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/fc1/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/fc1/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/fc2/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/fc2/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/final_layer_norm/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/final_layer_norm/scale": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/self_attn/k_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/self_attn/k_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/self_attn/out_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/self_attn/out_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/self_attn/q_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/self_attn/q_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/self_attn/v_proj/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/self_attn/v_proj/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/self_attn_layer_norm/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/46/self_attn_layer_norm/scale": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/47/fc1/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/47/fc1/kernel": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/47/fc2/bias": "flax_model-00009-of-00014.msgpack", | |
| "model/decoder/layers/47/fc2/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/final_layer_norm/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/final_layer_norm/scale": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/self_attn/k_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/self_attn/k_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/self_attn/out_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/self_attn/out_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/self_attn/q_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/self_attn/q_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/self_attn/v_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/self_attn/v_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/self_attn_layer_norm/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/47/self_attn_layer_norm/scale": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/fc1/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/fc1/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/fc2/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/fc2/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/final_layer_norm/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/final_layer_norm/scale": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/self_attn/k_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/self_attn/k_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/self_attn/out_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/self_attn/out_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/self_attn/q_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/self_attn/q_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/self_attn/v_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/self_attn/v_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/self_attn_layer_norm/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/48/self_attn_layer_norm/scale": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/fc1/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/fc1/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/fc2/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/fc2/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/final_layer_norm/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/final_layer_norm/scale": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/self_attn/k_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/self_attn/k_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/self_attn/out_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/self_attn/out_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/self_attn/q_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/self_attn/q_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/self_attn/v_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/self_attn/v_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/self_attn_layer_norm/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/49/self_attn_layer_norm/scale": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/fc1/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/fc1/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/fc2/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/fc2/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/final_layer_norm/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/final_layer_norm/scale": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/self_attn/k_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/self_attn/k_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/self_attn/out_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/self_attn/out_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/self_attn/q_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/self_attn/q_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/self_attn/v_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/self_attn/v_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/self_attn_layer_norm/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/5/self_attn_layer_norm/scale": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/fc1/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/fc1/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/fc2/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/fc2/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/final_layer_norm/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/final_layer_norm/scale": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/self_attn/k_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/self_attn/k_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/self_attn/out_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/self_attn/out_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/self_attn/q_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/self_attn/q_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/self_attn/v_proj/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/self_attn/v_proj/kernel": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/self_attn_layer_norm/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/50/self_attn_layer_norm/scale": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/51/fc1/bias": "flax_model-00010-of-00014.msgpack", | |
| "model/decoder/layers/51/fc1/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/fc2/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/fc2/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/final_layer_norm/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/final_layer_norm/scale": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/self_attn/k_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/self_attn/k_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/self_attn/out_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/self_attn/out_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/self_attn/q_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/self_attn/q_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/self_attn/v_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/self_attn/v_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/self_attn_layer_norm/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/51/self_attn_layer_norm/scale": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/fc1/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/fc1/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/fc2/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/fc2/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/final_layer_norm/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/final_layer_norm/scale": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/self_attn/k_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/self_attn/k_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/self_attn/out_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/self_attn/out_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/self_attn/q_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/self_attn/q_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/self_attn/v_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/self_attn/v_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/self_attn_layer_norm/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/52/self_attn_layer_norm/scale": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/fc1/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/fc1/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/fc2/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/fc2/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/final_layer_norm/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/final_layer_norm/scale": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/self_attn/k_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/self_attn/k_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/self_attn/out_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/self_attn/out_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/self_attn/q_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/self_attn/q_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/self_attn/v_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/self_attn/v_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/self_attn_layer_norm/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/53/self_attn_layer_norm/scale": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/fc1/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/fc1/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/fc2/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/fc2/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/final_layer_norm/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/final_layer_norm/scale": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/self_attn/k_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/self_attn/k_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/self_attn/out_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/self_attn/out_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/self_attn/q_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/self_attn/q_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/self_attn/v_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/self_attn/v_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/self_attn_layer_norm/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/54/self_attn_layer_norm/scale": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/fc1/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/fc1/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/fc2/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/fc2/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/final_layer_norm/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/final_layer_norm/scale": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/self_attn/k_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/self_attn/k_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/self_attn/out_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/self_attn/out_proj/kernel": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/self_attn/q_proj/bias": "flax_model-00011-of-00014.msgpack", | |
| "model/decoder/layers/55/self_attn/q_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/55/self_attn/v_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/55/self_attn/v_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/55/self_attn_layer_norm/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/55/self_attn_layer_norm/scale": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/fc1/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/fc1/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/fc2/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/fc2/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/final_layer_norm/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/final_layer_norm/scale": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/self_attn/k_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/self_attn/k_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/self_attn/out_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/self_attn/out_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/self_attn/q_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/self_attn/q_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/self_attn/v_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/self_attn/v_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/self_attn_layer_norm/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/56/self_attn_layer_norm/scale": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/fc1/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/fc1/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/fc2/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/fc2/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/final_layer_norm/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/final_layer_norm/scale": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/self_attn/k_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/self_attn/k_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/self_attn/out_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/self_attn/out_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/self_attn/q_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/self_attn/q_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/self_attn/v_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/self_attn/v_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/self_attn_layer_norm/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/57/self_attn_layer_norm/scale": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/fc1/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/fc1/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/fc2/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/fc2/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/final_layer_norm/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/final_layer_norm/scale": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/self_attn/k_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/self_attn/k_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/self_attn/out_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/self_attn/out_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/self_attn/q_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/self_attn/q_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/self_attn/v_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/self_attn/v_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/self_attn_layer_norm/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/58/self_attn_layer_norm/scale": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/fc1/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/fc1/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/fc2/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/fc2/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/final_layer_norm/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/final_layer_norm/scale": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/self_attn/k_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/self_attn/k_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/self_attn/out_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/self_attn/out_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/self_attn/q_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/self_attn/q_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/self_attn/v_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/self_attn/v_proj/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/self_attn_layer_norm/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/59/self_attn_layer_norm/scale": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/6/fc1/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/6/fc1/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/6/fc2/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/6/fc2/kernel": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/6/final_layer_norm/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/6/final_layer_norm/scale": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/6/self_attn/k_proj/bias": "flax_model-00012-of-00014.msgpack", | |
| "model/decoder/layers/6/self_attn/k_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/6/self_attn/out_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/6/self_attn/out_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/6/self_attn/q_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/6/self_attn/q_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/6/self_attn/v_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/6/self_attn/v_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/6/self_attn_layer_norm/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/6/self_attn_layer_norm/scale": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/fc1/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/fc1/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/fc2/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/fc2/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/final_layer_norm/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/final_layer_norm/scale": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/self_attn/k_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/self_attn/k_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/self_attn/out_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/self_attn/out_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/self_attn/q_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/self_attn/q_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/self_attn/v_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/self_attn/v_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/self_attn_layer_norm/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/60/self_attn_layer_norm/scale": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/fc1/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/fc1/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/fc2/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/fc2/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/final_layer_norm/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/final_layer_norm/scale": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/self_attn/k_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/self_attn/k_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/self_attn/out_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/self_attn/out_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/self_attn/q_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/self_attn/q_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/self_attn/v_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/self_attn/v_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/self_attn_layer_norm/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/61/self_attn_layer_norm/scale": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/fc1/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/fc1/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/fc2/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/fc2/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/final_layer_norm/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/final_layer_norm/scale": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/self_attn/k_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/self_attn/k_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/self_attn/out_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/self_attn/out_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/self_attn/q_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/self_attn/q_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/self_attn/v_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/self_attn/v_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/self_attn_layer_norm/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/62/self_attn_layer_norm/scale": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/fc1/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/fc1/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/fc2/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/fc2/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/final_layer_norm/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/final_layer_norm/scale": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/self_attn/k_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/self_attn/k_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/self_attn/out_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/self_attn/out_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/self_attn/q_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/self_attn/q_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/self_attn/v_proj/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/self_attn/v_proj/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/self_attn_layer_norm/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/63/self_attn_layer_norm/scale": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/7/fc1/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/7/fc1/kernel": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/7/fc2/bias": "flax_model-00013-of-00014.msgpack", | |
| "model/decoder/layers/7/fc2/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/final_layer_norm/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/final_layer_norm/scale": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/self_attn/k_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/self_attn/k_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/self_attn/out_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/self_attn/out_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/self_attn/q_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/self_attn/q_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/self_attn/v_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/self_attn/v_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/self_attn_layer_norm/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/7/self_attn_layer_norm/scale": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/fc1/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/fc1/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/fc2/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/fc2/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/final_layer_norm/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/final_layer_norm/scale": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/self_attn/k_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/self_attn/k_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/self_attn/out_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/self_attn/out_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/self_attn/q_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/self_attn/q_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/self_attn/v_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/self_attn/v_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/self_attn_layer_norm/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/8/self_attn_layer_norm/scale": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/fc1/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/fc1/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/fc2/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/fc2/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/final_layer_norm/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/final_layer_norm/scale": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/self_attn/k_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/self_attn/k_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/self_attn/out_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/self_attn/out_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/self_attn/q_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/self_attn/q_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/self_attn/v_proj/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/self_attn/v_proj/kernel": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/self_attn_layer_norm/bias": "flax_model-00014-of-00014.msgpack", | |
| "model/decoder/layers/9/self_attn_layer_norm/scale": "flax_model-00014-of-00014.msgpack" | |
| } | |
| } | |