INDEX

Explanations

mentions of the name "Karl" with varying levels of activation

oai_token-act-pair · gpt-3.5-turbo

the name "Karl."

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

flix

-0.77

ndra

-0.71

NX

-0.69

 Dangerous

-0.67

 Called

-0.66

nder

-0.64

LV

-0.63

 Doct

-0.62

Mub

-0.61

 Karma

-0.60

POSITIVE LOGITS

ounge

1.15

anguage

1.10

ophone

1.02

owship

1.00

ength

0.94

oths

0.91

ibrary

0.91

ottesville

0.91

otta

0.90

atan

0.89

Activations Density 0.019%

mentions of the name "Karl" with varying levels of activation

the name "Karl."

No Comments

No Known Activations