A collections of Gemma3n E4B models fintuned using the PIPPA and NPC-dialogue datasets, specifically aimed to be embedded in a game.
Hexi Wang PRO
chimbiwide
AI & ML interests
Fine-tuning small models for Roleplaying
Recent Activity
updated
a dataset
2 days ago
chimbiwide/brainstorming-thinking
new activity
2 days ago
chimbiwide/brainstorming-thinking:Upload brainstorming-thinking.csv
new activity
2 days ago
chimbiwide/sciqa-thinking:[bot] Conversion to Parquet
Organizations
A new attempt at training Gemma3NPC, with less conservative training parameters
Using the same dataset as Gemma3NPC-Filtered, but with a much higher learning rate and 3 epochs compared to 1.
A collection of all the GemmaReLe models, a special version of Gemma3NPC, made specifically for ReLe
A collections of Gemma3n E4B models fintuned using the pippadataset, aimed to be a general roleplaying model
A collections of Gemma3n E4B models fintuned using the pippa_filtered dataset, aimed to be a roleplaying model, though a pretty bad one
A collections of Gemma3n E4B models fintuned using the PIPPA and NPC-dialogue datasets, specifically aimed to be embedded in a game.
A collection of all the GemmaReLe models, a special version of Gemma3NPC, made specifically for ReLe
A new attempt at training Gemma3NPC, with less conservative training parameters
A collections of Gemma3n E4B models fintuned using the pippadataset, aimed to be a general roleplaying model
Using the same dataset as Gemma3NPC-Filtered, but with a much higher learning rate and 3 epochs compared to 1.
A collections of Gemma3n E4B models fintuned using the pippa_filtered dataset, aimed to be a roleplaying model, though a pretty bad one