INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Grind

-0.92

 Mahar

-0.74

 Mechdragon

-0.73

ensor

-0.71

ikan

-0.68

imon

-0.67

Bei

-0.64

isoft

-0.62

aband

-0.61

 spokes

-0.60

POSITIVE LOGITS

ctx

0.75

ARDS

0.74

zanne

0.73

swick

0.69

=]

0.68

otype

0.65

vision

0.65

 Jackets

0.65

Vs

0.64

$.

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.