INDEX

Explanations

years related to people or organizations being mentioned

oai_token-act-pair · gpt-3.5-turbo

the presence of significant names or titles associated with notable individuals or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

acebook

-0.84

olves

-0.83

interstitial

-0.77

ategory

-0.76

alde

-0.76

holes

-0.69

mint

-0.69

ertodd

-0.67

illion

-0.66

hole

-0.66

POSITIVE LOGITS

 supra

1.11

 Constantin

0.80

nee

0.79

et

0.79

Ala

0.75

LLP

0.73

 Wash

0.73

Rom

0.73

 Mata

0.72

 Colo

0.72

Activations Density 0.145%

years related to people or organizations being mentioned

the presence of significant names or titles associated with notable individuals or events

No Comments

No Known Activations