INDEX

Explanations

phrases or names related to the term "Re"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the abbreviation "Re," likely related to specific individuals or events involving someone named Re

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

NESS

-0.70

 Pebble

-0.67

DragonMagazine

-0.64

 handwriting

-0.63

 Korra

-0.59

avascript

-0.59

 Eagle

-0.59

 puzz

-0.59

 Icelandic

-0.58

 trem

-0.57

POSITIVE LOGITS

levant

1.30

usable

1.21

uben

1.21

asonable

1.18

gex

1.14

claim

1.14

versible

1.14

ceive

1.13

acher

1.12

clamation

1.12

Activations Density 0.022%

phrases or names related to the term "Re"

occurrences of the abbreviation "Re," likely related to specific individuals or events involving someone named Re

No Comments

No Known Activations