INDEX

Explanations

words related to physical proximity or scrutiny

oai_token-act-pair · gpt-3.5-turbo

phrases related to close observation and scrutiny

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

Bir

-0.68

Ame

-0.65

ĳå£«

-0.63

Pog

-0.62

 Palest

-0.61

 delinqu

-0.61

 Brav

-0.61

 Lerner

-0.60

 Nile

-0.59

 Herrera

-0.59

POSITIVE LOGITS

nery

0.81

ptives

0.80

alore

0.77

nered

0.73

aminer

0.72

atches

0.71

essional

0.70

apixel

0.69

piring

0.69

enment

0.69

Activations Density 0.093%

words related to physical proximity or scrutiny

phrases related to close observation and scrutiny

No Comments

No Known Activations