INDEX

Explanations

words related to being well-informed or making responsible decisions

oai_token-act-pair · gpt-3.5-turbo

terms related to informed decision-making and responsibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

stals

-0.83

estamp

-0.73

AMS

-0.72

bows

-0.70

 mysteriously

-0.68

 Devi

-0.68

nen

-0.67

oried

-0.66

yip

-0.65

Reloaded

-0.65

POSITIVE LOGITS

 observer

1.07

 consideration

1.06

 disclosure

1.00

 consent

0.97

 contemplation

0.96

 decision

0.96

 adherence

0.95

 stewards

0.93

ness

0.92

 bystand

0.92

Activations Density 0.133%

words related to being well-informed or making responsible decisions

terms related to informed decision-making and responsibility

No Comments

No Known Activations