INDEX

Explanations

phrases indicating an explanation or rephrasing of a concept

oai_token-act-pair · gpt-3.5-turbo

references to alternative explanations or interpretations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

yip

-0.79

iste

-0.66

atism

-0.65

achelor

-0.65

 outweigh

-0.62

 overcame

-0.62

acebook

-0.61

ttle

-0.58

apego

-0.57

ousel

-0.56

POSITIVE LOGITS

 words

1.25

words

1.18

worldly

1.11

 respects

0.97

 contexts

0.93

 instances

0.86

 circumstances

0.84

parts

0.84

 parts

0.84

word

0.82

Activations Density 0.033%

phrases indicating an explanation or rephrasing of a concept

references to alternative explanations or interpretations

No Comments

No Known Activations