rasa/cmd_gen_codellama_13b_calm_demo
Full fine-tune vs codellama/CodeLlama-13b-Instruct-hf. Fine-tune: 97% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 0/1, Norms 7/81, MLP 120/120, Attention 160/160, lm_head 0/1
Full fine-tune vs codellama/CodeLlama-13b-Instruct-hf. Fine-tune: 97% of the weights changed, across 40 of 40 layers.
Tensors changed: Embeddings 0/1, Norms 7/81, MLP 120/120, Attention 160/160, lm_head 0/1