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

orneys

-0.76

andal

-0.74

mble

-0.73

ibilities

-0.68

igger

-0.66

umbing

-0.65

undown

-0.63

hog

-0.63

 ambush

-0.62

ornings

-0.62

POSITIVE LOGITS

thro

0.80

 Hath

0.72

emort

0.68

Lap

0.67

Plex

0.66

 Eisen

0.66

ãĥĥãĥī

0.66

Ult

0.65

sold

0.64

ãĥķãĤ©

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.