Model assets for the first Mixture-of-Lora technique applied to Llama. https://bit.ly/48bqshl
crumb
crumb
AI & ML interests
None yet
Recent Activity
updated
a collection
about 5 hours ago
CLM_R
updated
a dataset
about 11 hours ago
crumbs-playground/loss-balancing-clmr-4B-dora-rb256-cb256-bs8-lr2e-05-gn0.5-eb0.01
published
a dataset
about 11 hours ago
crumbs-playground/loss-balancing-clmr-4B-dora-rb256-cb256-bs8-lr2e-05-gn0.5-eb0.01
Organizations
GPT2 Models using Linear layers instead of Conv layers for convenience.
Smaller models trained locally on my 2xA6000 Lambda Vector
Model assets for the first Mixture-of-Lora technique applied to Llama. https://bit.ly/48bqshl
First Prototype of the second iteration of MoLora utilizing mixture of expert techniques applied to the Llama2 model.
GPT2 Models using Linear layers instead of Conv layers for convenience.
Parts of Meta's LlamaV2 models, chopped up and trained. CoreX means the first X layers were kept.
Smaller models trained locally on my 2xA6000 Lambda Vector