INDEX

Explanations

words related to sequels or subsequent installments in a series

oai_token-act-pair · gpt-3.5-turbo

mentions of film sequels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hoff

-0.63

athering

-0.63

 Athletics

-0.62

¬¼

-0.62

·

-0.60

aching

-0.59

romy

-0.59

 Organization

-0.58

WI

-0.57

 Hockey

-0.57

POSITIVE LOGITS

 sequels

1.08

 sequel

1.07

arium

0.91

 trilogy

0.86

ngth

0.84

arily

0.76

 thereto

0.76

ogue

0.71

 playthrough

0.69

 installments

0.69

Activations Density 0.011%

words related to sequels or subsequent installments in a series

mentions of film sequels

No Comments

No Known Activations