INDEX

Explanations

phrases related to negative consequences

oai_token-act-pair · gpt-3.5-turbo

instances of the word "the."

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

pez

-0.76

å§«

-0.73

razil

-0.72

advertising

-0.71

cember

-0.68

LIN

-0.66

Notes

-0.66

çīĪ

-0.65

LAB

-0.65

é¾įåĸļå£«

-0.65

POSITIVE LOGITS

 slightest

1.31

 brightest

1.06

 usual

0.96

 smartest

0.94

 easiest

0.94

 same

0.91

 prett

0.90

 strongest

0.86

 stereotypical

0.84

 norm

0.83

Activations Density 0.068%

phrases related to negative consequences

instances of the word "the."

No Comments

No Known Activations