INDEX

Explanations

words related to different groups of people and their roles or characteristics

oai_token-act-pair · gpt-3.5-turbo

references to various groups of people or professions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

DonaldTrump

-0.77

BALL

-0.66

forward

-0.65

ield

-0.62

ģ«

-0.61

ray

-0.60

paragraph

-0.58

VIEW

-0.58

UTERS

-0.58

dain

-0.57

POSITIVE LOGITS

folk

1.20

 themselves

0.97

0.86

hest

0.84

iest

0.83

']

0.80

layer

0.80

 involved

0.74

heet

0.73

hip

0.70

Activations Density 0.231%

words related to different groups of people and their roles or characteristics

references to various groups of people or professions

No Comments

No Known Activations