INDEX

Explanations

phrases containing the word "nothing" followed by the phrase "more" and a verb

oai_token-act-pair · gpt-3.5-turbo

phrases indicating a sense of negation or minimalism in expression

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

ahime

-0.90

[|

-0.71

church

-0.68

 Pengu

-0.63

aii

-0.63

ivas

-0.61

sten

-0.60

visor

-0.60

ppel

-0.60

 Elements

-0.59

POSITIVE LOGITS

 except

0.96

except

0.84

 whatsoever

0.83

 transpired

0.73

pires

0.72

 resembling

0.72

Except

0.70

 besides

0.69

 happening

0.66

 stopping

0.58

Activations Density 0.166%

phrases containing the word "nothing" followed by the phrase "more" and a verb

phrases indicating a sense of negation or minimalism in expression

No Comments

No Known Activations