INDEX

Explanations

references to works of fiction or literary genres

oai_token-act-pair · gpt-3.5-turbo

references to fiction, particularly science fiction and its subgenres

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

umm

-0.71

arov

-0.68

Ķ

-0.65

xon

-0.64

 Parenthood

-0.64

realDonaldTrump

-0.64

Accessory

-0.63

ermanent

-0.61

lain

-0.61

 downed

-0.60

POSITIVE LOGITS

 fiction

1.06

 Fiction

0.95

fiction

0.93

 novels

0.83

 imagin

0.80

 anthology

0.80

 Writers

0.78

 novelist

0.76

Writ

0.76

reading

0.75

Activations Density 0.013%

references to works of fiction or literary genres

references to fiction, particularly science fiction and its subgenres

No Comments

No Known Activations