INDEX

Explanations

the phrase "believe it or not."

oai_token-act-pair · gpt-3.5-turbo

phrases that express doubt or disbelief

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

istries

-0.64

culosis

-0.59

RH

-0.58

 Grac

-0.57

 Rollins

-0.56

nen

-0.56

HIP

-0.56

restling

-0.55

OU

-0.55

pload

-0.55

POSITIVE LOGITS

 versa

0.83

 thereof

0.83

éĹ

0.80

.}

0.70

cffff

0.67

cffffcc

0.65

endif

0.65

ALSE

0.64

 alike

0.63

depending

0.63

Activations Density 0.097%

the phrase "believe it or not."

phrases that express doubt or disbelief

No Comments

No Known Activations