INDEX

Explanations

phrases indicating a need for action or instruction

oai_token-act-pair · gpt-3.5-turbo

phrases emphasizing the word "have."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

AH

-0.71

icy

-0.71

estic

-0.68

weed

-0.66

Discuss

-0.65

adjusted

-0.65

emis

-0.64

 Returning

-0.63

act

-0.62

igraph

-0.61

POSITIVE LOGITS

 yourselves

0.88

 yourself

0.78

Tube

0.78

 gotta

0.74

 mileage

0.71

issance

0.68

imar

0.68

 shalt

0.68

 doubtless

0.67

 hereby

0.67

Activations Density 0.174%

phrases indicating a need for action or instruction

phrases emphasizing the word "have."

No Comments

No Known Activations