INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

nil

-0.73

Ru

-0.64

issance

-0.63

eb

-0.63

prev

-0.63

ucl

-0.62

asonable

-0.62

XP

-0.62

obin

-0.61

eful

-0.61

POSITIVE LOGITS

noon

0.74

 Tanz

0.72

halla

0.66

 Liang

0.66

orpor

0.64

 Galile

0.61

Kau

0.59

isk

0.59

ctica

0.59

 supper

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.