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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Tradable

-0.87

eed

-0.82

untled

-0.81

 Cricket

-0.76

ament

-0.75

ioned

-0.74

ariat

-0.74

uid

-0.72

ItemTracker

-0.69

Thumbnail

-0.69

POSITIVE LOGITS

pal

0.68

ï

0.67

terday

0.65

knife

0.65

banks

0.63

please

0.63

south

0.62

iHUD

0.62

ston

0.61

Mor

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.