INDEX

Explanations

mentions of the word "Euro" alongside numbers

oai_token-act-pair · gpt-3.5-turbo

references to European events or organizations

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

INESS

-0.77

OHN

-0.76

enegger

-0.74

////////////////

-0.73

hips

-0.73

IAL

-0.72

âĶĢ

-0.72

::::::::

-0.71

à©

-0.70

âĶĢâĶĢ

-0.68

POSITIVE LOGITS

pol

1.06

vision

0.95

pes

0.92

pa

0.80

netic

0.79

stat

0.79

 Euro

0.78

pay

0.78

centric

0.78

cop

0.77

Activations Density 0.007%

mentions of the word "Euro" alongside numbers

references to European events or organizations

No Comments

No Known Activations