INDEX

Explanations

phrases or words containing 'ew'

oai_token-act-pair · gpt-3.5-turbo

instances of a specific term related to a phenomenon or concept

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

REDACTED

-0.66

 cort

-0.65

 unarmed

-0.65

 suspic

-0.64

 administr

-0.59

 anat

-0.58

 retri

-0.58

 inhibition

-0.58

Downloadha

-0.58

 apprehension

-0.57

POSITIVE LOGITS

estern

1.19

een

1.17

riter

1.07

esley

1.04

esome

1.04

ITNESS

1.03

sburg

0.97

alker

0.93

ritten

0.92

eh

0.91

Activations Density 0.016%

phrases or words containing 'ew'

instances of a specific term related to a phenomenon or concept

No Comments

No Known Activations