INDEX

Explanations

words related to individuals' names or identities

oai_token-act-pair · gpt-3.5-turbo

characters' names and proper nouns in the context of specific narratives or 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

nown

-0.67

ãĤ¶

-0.66

ulhu

-0.65

lehem

-0.64

»Ĵ

-0.63

userc

-0.62

runtime

-0.62

taboola

-0.60

URA

-0.60

natureconservancy

-0.60

POSITIVE LOGITS

backer

0.99

ordan

0.76

 cuff

0.73

CTR

0.72

kefeller

0.71

 flats

0.70

restling

0.68

earchers

0.67

kowski

0.66

 McMaster

0.66

Activations Density 0.596%

words related to individuals' names or identities

characters' names and proper nouns in the context of specific narratives or events

No Comments

No Known Activations