|
Download README.md from CodeDevX/auralis-coder-dataset: direct link, hf CLI and curl.
- Browser
- Download file 1.2 kB
-
https://huggingface.co/datasets/CodeDevX/auralis-coder-dataset/resolve/main/README.md
- Command line
-
hf download hf://datasets/CodeDevX/auralis-coder-dataset/README.md
-
curl -L -o README.md https://huggingface.co/datasets/CodeDevX/auralis-coder-dataset/resolve/main/README.md
1.2 kB
| license: mit | |
| language: | |
| - en | |
| pretty_name: a | |
| # Auralis-Coder Training Dataset | |
| Packed token dataset for training the **Auralis-Coder** 1.5B parameter language model. | |
| ## Dataset Overview | |
| | Property | Value | | |
| |---|---| | |
| | Total tokens | 176,544,387 | | |
| | Tokenizer | Qwen2.5 (Qwen/Qwen2.5-1.5B) | | |
| | Format | uint32 token IDs (packed) | | |
| | Shards | 16 | | |
| | Shard size | ~10M tokens each | | |
| | Documents | 235,219 | | |
| | Duplicates removed | 19 | | |
| ## Tokenization & Packing | |
| - Tokenized with Qwen2.5 tokenizer (vocab: 151,936) | |
| - Packed into fixed-size shards (10M tokens each) | |
| - Sequence length: 4,096 tokens during training | |
| - Maximum context: 131,072 tokens | |
| ## Usage | |
| ```python | |
| import numpy as np | |
| # Load a shard | |
| tokens = np.fromfile("tokens-000000.bin", dtype=np.uint32) | |
| # Reshape into sequences (e.g., 4096 tokens per sequence) | |
| seq_len = 4096 | |
| sequences = tokens.reshape(-1, seq_len) | |
| ``` | |
| ## Files | |
| - `tokens-000000.bin` through `tokens-000015.bin` — Packed token shards (uint32) | |
| - `packed_manifest.json` — Dataset statistics and metadata | |
| - `dataset.json` — Additional metadata | |
| ## License | |
| Refer to the source dataset licenses (FineWeb, StarCoder, FineWeb-Edu) and the Auralis project distribution. |