INDEX

Explanations

the word "a"

oai_token-act-pair · gpt-3.5-turbo

indicators related to frequency or occurrence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

ubs

-0.78

chu

-0.65

chnology

-0.64

anguage

-0.64

 Class

-0.63

IFF

-0.63

cler

-0.61

 Benedict

-0.60

arest

-0.59

ammy

-0.59

POSITIVE LOGITS

ItemTracker

0.78

 Canaveral

0.77

 partName

0.74

effic

0.73

borgh

0.72

ayne

0.71

éĹĺ

0.69

saf

0.66

EStreamFrame

0.65

 Flavoring

0.65

Activations Density 0.000%

the word "a"

indicators related to frequency or occurrence

No Comments

No Known Activations