INDEX

Explanations

words containing "art."

oai_token-act-pair · gpt-3.5-turbo

references to art and artistic concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Crosby

-0.66

 Aval

-0.65

ELD

-0.63

 chloride

-0.61

 cumbers

-0.61

 mainland

-0.59

 snowy

-0.59

 Isles

-0.58

 crest

-0.58

 sclerosis

-0.57

POSITIVE LOGITS

ifact

1.31

isans

1.16

heid

1.15

illery

1.14

ooth

1.04

ificial

1.04

ners

1.03

ICLE

1.02

ner

1.00

inez

0.98

Activations Density 0.029%

words containing "art."

references to art and artistic concepts

No Comments

No Known Activations