INDEX

Explanations

threshold values, such as specific levels of income or measurements

oai_token-act-pair · gpt-3.5-turbo

comparisons involving numerical values or thresholds

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

ires

-0.73

Leaks

-0.68

 Pony

-0.66

parency

-0.65

rodu

-0.65

 Reporter

-0.64

edia

-0.64

ocene

-0.64

Beh

-0.63

eval

-0.62

POSITIVE LOGITS

ifice

1.21

acle

1.21

ific

1.16

chard

1.11

acles

1.03

nam

0.99

acular

0.97

phans

0.96

chid

0.95

lando

0.95

Activations Density 0.098%

threshold values, such as specific levels of income or measurements

comparisons involving numerical values or thresholds

No Comments

No Known Activations