INDEX

Explanations

punctuation marks, specifically commas

oai_token-act-pair · gpt-3.5-turbo

phrases related to alternatives or choices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ĸļ

-0.70

APTER

-0.64

atform

-0.64

VERTISEMENT

-0.60

bery

-0.60

reach

-0.57

ertodd

-0.57

 ethic

-0.56

 pores

-0.56

orage

-0.55

POSITIVE LOGITS

Else

1.06

abouts

1.03

 alternatively

1.02

acle

0.86

acles

0.83

thodox

0.81

depending

0.80

phans

0.77

 equival

0.75

matically

0.74

Activations Density 0.106%

punctuation marks, specifically commas

phrases related to alternatives or choices

No Comments

No Known Activations