INDEX

Explanations

days of the week, months, or specific dates mentioned in the text

oai_token-act-pair · gpt-3.5-turbo

phrases indicating specific days of the week

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

andel

-0.71

ANY

-0.71

GBT

-0.66

netflix

-0.66

agos

-0.66

Limit

-0.65

common

-0.65

erate

-0.64

deal

-0.64

Common

-0.64

POSITIVE LOGITS

pires

1.03

 nutshell

0.88

pired

0.70

eatures

0.69

 airs

0.65

Iv

0.61

ked

0.61

 Flickr

0.59

 Responsibility

0.59

ÃĥÃĤ

0.58

Activations Density 0.440%

days of the week, months, or specific dates mentioned in the text

phrases indicating specific days of the week

No Comments

No Known Activations