INDEX

Explanations

numerical values perceived as significant or important

oai_token-act-pair · gpt-3.5-turbo

numerical values and their representations in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ItemImage

-0.87

liga

-0.74

lehem

-0.70

cerning

-0.69

ItemThumbnailImage

-0.68

ercise

-0.67

league

-0.65

holding

-0.64

Lago

-0.63

£ı

-0.62

POSITIVE LOGITS

xff

1.06

1.01

xd

0.89

xe

0.86

0.85

xb

0.80

Â°

0.77

xc

0.77

xa

0.73

0.71

Activations Density 0.015%

numerical values perceived as significant or important

numerical values and their representations in various contexts

No Comments

No Known Activations