INDEX

Explanations

the name "Kat" with varying levels of activation

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Kat" in various contexts

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

cffff

-0.86

LEASE

-0.80

CLASSIFIED

-0.79

uploads

-0.74

prise

-0.72

vous

-0.67

interstitial

-0.67

lict

-0.67

 confir

-0.67

 margins

-0.66

POSITIVE LOGITS

Kat

1.12

anas

1.12

kat

0.95

usha

0.94

Kat

0.91

apesh

0.90

anz

0.87

inka

0.85

ney

0.85

owitz

0.85

Activations Density 0.011%

the name "Kat" with varying levels of activation

mentions of the name "Kat" in various contexts

No Comments

No Known Activations