# bge-small-zh-v1.5

Hugging Face: `BAAI/bge-small-zh-v1.5`
Reconstructed from `config.json` alone. No weights were read.

| | |
|---|---|
| Derived from structure | 23,685,120 |
| In the published checkpoint | 23,954,432 (safetensors.total, 2023-10-12) |
| Delta | -1.12% |
| Comparability | comparable |
| Layers | 18 |
| Priced on | A10G (24GB) |
| Est. one run | $0.30 |

## Structure

| # | Layer | Type | Output shape |
|---|---|---|---|
| 1 | Input | Input | 1 × 512 |
| 2 | Embedding | Embedding | 1 × 512 × 512 |
| 3 | Positional_Embedding | Learned Pos Embed | 1 × 512 × 512 |
| 4 | Attention_1 | Multi-Head Attention | 1 × 512 × 512 |
| 5 | Add_1 | Add | 1 × 512 × 512 |
| 6 | LayerNorm_1_1 | LayerNorm | 1 × 512 × 512 |
| 7 | FFN_1 | Feed Forward | 1 × 512 × 512 |
| 8 | Attention_2 | Multi-Head Attention | 1 × 512 × 512 |
| 9 | Add_2 | Add | 1 × 512 × 512 |
| 10 | LayerNorm_2_1 | LayerNorm | 1 × 512 × 512 |
| 11 | FFN_2 | Feed Forward | 1 × 512 × 512 |
| 12 | Attention_3 | Multi-Head Attention | 1 × 512 × 512 |
| 13 | Add_3 | Add | 1 × 512 × 512 |
| 14 | LayerNorm_3_1 | LayerNorm | 1 × 512 × 512 |
| 15 | FFN_3 | Feed Forward | 1 × 512 × 512 |
| 16 | Attention_4 | Multi-Head Attention | 1 × 512 × 512 |
| 17 | Add_4 | Add | 1 × 512 × 512 |
| 18 | LayerNorm_4_1 | LayerNorm | 1 × 512 × 512 |
| 19 | FFN_4 | Feed Forward | 1 × 512 × 512 |
| 20 | Output | Output | 1 × 512 × 512 |

## Do this to your own model

```
pip install neurarch-trace
neurarch-trace BAAI/bge-small-zh-v1.5 --plan --share
```

- Page: https://neurarch.com/m/bge-small-zh-v1-5.html
- All models: https://neurarch.com/m/index.json
- The checks: https://neurarch.com/r/index.json