INDEX

Explanations

phrases containing the words "no one"

oai_token-act-pair · gpt-3.5-turbo

the repetition of the word "one."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

anga

-0.63

prem

-0.62

xton

-0.60

ypes

-0.59

efully

-0.59

gur

-0.58

 KING

-0.56

utterstock

-0.56

ahime

-0.56

 Territories

-0.55

POSITIVE LOGITS

 else

1.13

 whatsoever

0.90

 dime

0.89

 bothered

0.81

ody

0.76

rieve

0.72

 cared

0.72

 imaginable

0.72

else

0.71

 answ

0.71

Activations Density 0.031%

phrases containing the words "no one"

the repetition of the word "one."

No Comments

No Known Activations