INDEX

Explanations

events or activities that are starting or commencing

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the initiation of events

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

olog

-0.75

enne

-0.69

avering

-0.69

é¾įå

-0.66

Pers

-0.64

WARE

-0.64

ulous

-0.64

ollen

-0.62

chn

-0.62

atial

-0.61

POSITIVE LOGITS

uberty

0.75

 puberty

0.74

nings

0.71

 kickoff

0.71

loading

0.70

stage

0.64

 Tinder

0.64

 takeoff

0.64

 tonight

0.63

 tomorrow

0.62

Activations Density 0.019%

events or activities that are starting or commencing

phrases indicating the initiation of events

No Comments

No Known Activations