Kronos BTC 1hr Tokenizer

This is a fine-tuned Kronos tokenizer trained on Bitcoin 1-hour candlestick data.

Model Details

  • Model Type: Kronos Tokenizer
  • Training Data: Bitcoin 1-hour candlestick data
  • Architecture: Transformer-based encoder-decoder
  • Model Size: ~15MB

Configuration

{
  "attn_dropout_p": 0.0,
  "beta": 0.05,
  "d_in": 6,
  "d_model": 256,
  "ff_dim": 512,
  "ffn_dropout_p": 0.0,
  "gamma": 1.1,
  "gamma0": 1.0,
  "group_size": 4,
  "n_dec_layers": 4,
  "n_enc_layers": 4,
  "n_heads": 4,
  "resid_dropout_p": 0.0,
  "s1_bits": 10,
  "s2_bits": 10,
  "zeta": 0.05
}

Usage

This tokenizer is part of the Kronos financial prediction system. It should be used in conjunction with the corresponding basemodel.

Training Details

  • Training Time: ~78 minutes
  • Framework: PyTorch

Files

  • config.json: Tokenizer configuration
  • model.safetensors: Tokenizer weights
  • README.md: This file

License

MIT License

Downloads last month
-
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support