Instructions to use mlx-community/ZONOS3-Mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- MLX
How to use mlx-community/ZONOS3-Mini with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] hf download mlx-community/ZONOS3-Mini --local-dir ZONOS3-Mini
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
- Atomic Chat
Download speaker_encoder/config.json from mlx-community/ZONOS3-Mini: direct link, hf CLI and curl.
- Browser
- Download file 123 kB
-
https://huggingface.co/mlx-community/ZONOS3-Mini/resolve/main/speaker_encoder/config.json
- Command line
-
hf download hf://mlx-community/ZONOS3-Mini/speaker_encoder/config.json
-
curl -L -o config.json https://huggingface.co/mlx-community/ZONOS3-Mini/resolve/main/speaker_encoder/config.json
123 kB
| { | |
| "metadata": { | |
| "GGUF.tensor_count": 913, | |
| "GGUF.kv_count": 19, | |
| "general.architecture": "mect_spk", | |
| "general.name": "ZONOS3 Mini speaker encoder ownmect_s2 (MECT-B2 + ASTP, 192-d)", | |
| "mect_spk.sample_rate": 16000, | |
| "mect_spk.max_seconds": 60, | |
| "mect_spk.embedding_length": 192, | |
| "mect_spk.n_mels": 80, | |
| "mect_spk.frame_length": 400, | |
| "mect_spk.frame_shift": 160, | |
| "mect_spk.n_fft": 512, | |
| "mect_spk.n_heads": 4, | |
| "mect_spk.n_experts": 4, | |
| "mect_spk.pos_kernel": 113, | |
| "mect_spk.preemphasis": 0.9700000286102295, | |
| "mect_spk.waveform_scale": 32768.0, | |
| "mect_spk.log_floor": 1.1920928955078125e-07, | |
| "mect_spk.ln_eps": 9.999999974752427e-07, | |
| "mect_spk.pool_var_floor": 1.0000000116860974e-07, | |
| "mect_spk.backbone": "MECT_B2", | |
| "mect_spk.source_tensor_count": 1253 | |
| }, | |
| "tensors": { | |
| "frontend.window": { | |
| "shape": [ | |
| 400 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "frontend.mel_banks": { | |
| "shape": [ | |
| 80, | |
| 257 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "frontend.resample_44100": { | |
| "shape": [ | |
| 160, | |
| 475 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.head.conv.weight": { | |
| "shape": [ | |
| 32, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.head.conv.bias": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.conv2.w2": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.conv2.w3": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.sc.weight": { | |
| "shape": [ | |
| 128, | |
| 32, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.reduce.weight": { | |
| "shape": [ | |
| 32, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.reduce.bias": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.conv2.w2": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.conv2.w3": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.sc.weight": { | |
| "shape": [ | |
| 128, | |
| 32, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.reduce.weight": { | |
| "shape": [ | |
| 32, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.reduce.bias": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.conv2.w2": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.conv2.w3": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.sc.weight": { | |
| "shape": [ | |
| 128, | |
| 32, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.reduce.weight": { | |
| "shape": [ | |
| 64, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.reduce.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.sc.weight": { | |
| "shape": [ | |
| 128, | |
| 64, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.reduce.weight": { | |
| "shape": [ | |
| 64, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.reduce.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.sc.weight": { | |
| "shape": [ | |
| 128, | |
| 64, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.reduce.weight": { | |
| "shape": [ | |
| 64, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.reduce.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.sc.weight": { | |
| "shape": [ | |
| 128, | |
| 64, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 128, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 128, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.reduce.weight": { | |
| "shape": [ | |
| 64, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.reduce.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.conv2.w2": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.conv2.w3": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.sc.weight": { | |
| "shape": [ | |
| 256, | |
| 64, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.reduce.weight": { | |
| "shape": [ | |
| 128, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.reduce.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.sc.weight": { | |
| "shape": [ | |
| 256, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.reduce.weight": { | |
| "shape": [ | |
| 128, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.reduce.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.sc.weight": { | |
| "shape": [ | |
| 256, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.reduce.weight": { | |
| "shape": [ | |
| 128, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.reduce.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.sc.weight": { | |
| "shape": [ | |
| 256, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.reduce.weight": { | |
| "shape": [ | |
| 128, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.reduce.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.sc.weight": { | |
| "shape": [ | |
| 256, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.reduce.weight": { | |
| "shape": [ | |
| 128, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.reduce.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.conv2.w1": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.sc.weight": { | |
| "shape": [ | |
| 256, | |
| 128, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.2.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.2.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.2.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.2.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.2.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.2.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.sc.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.sc.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.1.conv1.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.1.pw1.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.1.pw1.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.1.conv2.w0": { | |
| "shape": [ | |
| 256, | |
| 1, | |
| 3, | |
| 3 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.1.pw2.weight": { | |
| "shape": [ | |
| 256, | |
| 256, | |
| 1, | |
| 1 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.1.pw2.bias": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.rproj.weight": { | |
| "shape": [ | |
| 64, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.rproj.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.rproj.ln.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.rproj.ln.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.pos.dw.weight": { | |
| "shape": [ | |
| 64, | |
| 113 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.pos.dw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.pos.pw.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.pos.pw.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.attn.q.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.attn.q.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.attn.k.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.attn.k.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.attn.v.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.attn.v.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.attn.o.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.attn.o.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.ln1.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.ln1.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.moe.gate.weight": { | |
| "shape": [ | |
| 4, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.moe.experts.weight": { | |
| "shape": [ | |
| 256, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.ffn_out.weight": { | |
| "shape": [ | |
| 64, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.ffn_out.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.ln2.weight": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.ln2.bias": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.eproj.weight": { | |
| "shape": [ | |
| 2560, | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.eproj.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.dense_w": { | |
| "shape": [ | |
| 4, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "pool.l1.weight": { | |
| "shape": [ | |
| 128, | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "pool.l1.bias": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "pool.l2.weight": { | |
| "shape": [ | |
| 2560, | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "pool.l2.bias": { | |
| "shape": [ | |
| 2560 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "linear.weight": { | |
| "shape": [ | |
| 192, | |
| 5120 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.head.bn.scale": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.head.bn.shift": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.reduce.bn.scale": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.reduce.bn.shift": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.0.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.reduce.bn.scale": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.reduce.bn.shift": { | |
| "shape": [ | |
| 32 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b1.1.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.reduce.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.reduce.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.0.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.reduce.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.reduce.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.1.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.reduce.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.reduce.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.2.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.reduce.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.reduce.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b2.3.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.reduce.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.reduce.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.0.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.reduce.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.reduce.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.1.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.reduce.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.reduce.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.2.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.reduce.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.reduce.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.3.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.reduce.bn.scale": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.reduce.bn.shift": { | |
| "shape": [ | |
| 128 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b3.4.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.2.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.2.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.2.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.cnn.2.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.0.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.1.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.sc.bn.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.0.sc.bn.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.1.bn1.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.1.bn1.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.1.bn2.scale": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.cnn.1.bn2.shift": { | |
| "shape": [ | |
| 256 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.pos.bn.scale": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "fe.b4.2.pos.bn.shift": { | |
| "shape": [ | |
| 64 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "bn.scale": { | |
| "shape": [ | |
| 5120 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| }, | |
| "bn.shift": { | |
| "shape": [ | |
| 5120 | |
| ], | |
| "matrix": false, | |
| "dtype": "float32" | |
| } | |
| } | |
| } | |