INDEX
Explanations
okay or alright
phrases that mark assistant turn-taking and conversational preambles, signaling the model’s persona and setup for a response in chat-style dialogue.
New Auto-Interp
Negative Logits
nutritive
0.27
fertilizers
0.27
pollinators
0.27
abiotic
0.26
ATMs
0.26
toxins
0.26
pointers
0.26
antioxidants
0.26
alleles
0.26
legis
0.26
POSITIVE LOGITS
jsem
0.44
我现在
0.41
我
0.40
please
0.37
আমি
0.36
我已经
0.35
Please
0.35
मैंने
0.35
Hello
0.35
मैंने
0.35
Activations Density 1.756%