|
--- |
|
language: en |
|
license: mit |
|
pipeline_tag: text-generation |
|
--- |
|
|
|
# **UPDATE (2023-09-23):** |
|
This model is obsolete. Thanks to quantization you can run AI Dungeon 2 Classic (a 1.5B model) under equivalent hardware. [See here](https://huggingface.co/Crataco/ggml-ai-dungeon-2-classic). |
|
*** |
|
# AID-Neo-125M |
|
|
|
## Model description |
|
This model was inspired by -- and finetuned on the same dataset of -- [KoboldAI's GPT-Neo-125M-AID (Mia) model](https://huggingface.co/KoboldAI/GPT-Neo-125M-AID): the AI Dungeon dataset (`text_adventures.txt`). This was to fix a possible oversight in the original model, which was trained with [an unfortunate bug](https://github.com/EricFillion/happy-transformer/issues/283). You could technically consider it a "retraining" of the same model using different software. |
|
|
|
### Training hyperparameters |
|
|
|
The following hyperparameters were used during training: |
|
- learning_rate: 5e-05 |
|
- train_batch_size: 1 |
|
- eval_batch_size: 1 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 3.0 |
