Dongximing

Question in fasttext

1

Hi Ben In fasttext, we can use bi-gram to get tokens, for example, I love your --> in bi-gram--> I, love, u, I love, love u. so if we set...

for word embedding in RNN model

2

Hi model.embedding.weight.data.copy_ is a fine-tuning way ? so is that mean they are the same as model parameters to train? I also find .pre_train to load word embedding, which means...

Hi everyone Did you face this problem? I just fixed the overfloat problem.

what is the meaning of contribution?

1

Hi everyone, I have a question about LLM contribution, in the picture. this is Perturbation-based Attribution method. the basic idea is that replace the words in order for example `I...

eval context

1

Hi, I was confused about the context of "the evaluation model input pairs", for me, I use deepseek-r1 and the model output has two parts one is "thinking" and the...