INDEX

Explanations

references to controversial or sensitive topics often related to political discussions

oai_token-act-pair · gpt-3.5-turbo

phrases indicating agreement or acknowledgment in a conversation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

idden

-0.77

tex

-0.74

pedia

-0.74

ccording

-0.74

idas

-0.72

iti

-0.72

Dialog

-0.72

vert

-0.72

SPONSORED

-0.70

Vert

-0.69

POSITIVE LOGITS

 caveats

0.93

 caveat

0.90

 exceptions

0.74

 plenty

0.73

 wouldn

0.72

 disclaim

0.66

 technically

0.65

 begs

0.65

 assumes

0.64

 none

0.63

Activations Density 0.135%

references to controversial or sensitive topics often related to political discussions

phrases indicating agreement or acknowledgment in a conversation

No Comments

No Known Activations