INDEX

Explanations

words related to transitioning to new topics or points in a discussion

oai_token-act-pair · gpt-3.5-turbo

phrases that signal transitions or connections between ideas

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 intimidated

-0.66

 tolerated

-0.63

 intimid

-0.63

 endors

-0.62

 destro

-0.58

 outnumbered

-0.57

cats

-0.57

soever

-0.56

 discriminate

-0.56

 exhibited

-0.55

POSITIVE LOGITS

amaz

0.85

 Conclusion

0.77

forth

0.76

 wondering

0.75

 QUEST

0.71

 conclude

0.71

zsche

0.70

raq

0.70

mber

0.69

oops

0.68

Activations Density 0.141%

words related to transitioning to new topics or points in a discussion

phrases that signal transitions or connections between ideas

No Comments

No Known Activations