INDEX

Explanations

phrases related to self-definition and self-description

oai_token-act-pair · gpt-3.5-turbo

expressions of opinion about individuals

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

teness

-0.80

 Dism

-0.65

 shock

-0.65

Kot

-0.63

 Extend

-0.63

 Indust

-0.60

win

-0.60

irm

-0.59

 Awakens

-0.59

 Ship

-0.59

POSITIVE LOGITS

 supposed

0.86

Interstitial

0.82

 happening

0.81

 gonna

0.77

nt

0.76

 going

0.75

 akin

0.74

 destined

0.74

omorphic

0.73

alian

0.71

Activations Density 0.203%

phrases related to self-definition and self-description

expressions of opinion about individuals

No Comments

No Known Activations