INDEX

Explanations

phrases related to making choices or trade-offs

oai_token-act-pair · gpt-3.5-turbo

terms related to trade-offs and evaluations of value

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

bered

-0.74

urses

-0.73

arus

-0.73

late

-0.71

liter

-0.69

miah

-0.68

bill

-0.68

attery

-0.65

bus

-0.65

estic

-0.65

POSITIVE LOGITS

 downside

0.92

why

0.86

 Problem

0.83

why

0.83

WHY

0.79

 weaknesses

0.78

 lesson

0.78

Problem

0.76

 drawback

0.76

 takeaway

0.75

Activations Density 0.531%

phrases related to making choices or trade-offs

terms related to trade-offs and evaluations of value

No Comments

No Known Activations