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

 payoff

-0.69

icum

-0.66

 PowerPoint

-0.63

 Idaho

-0.63

growth

-0.63

rose

-0.63

endum

-0.61

rition

-0.61

 yield

-0.60

gradation

-0.60

POSITIVE LOGITS

pport

0.88

Mos

0.72

ãĥ³ãĤ¸

0.72

Ø³

0.71

acht

0.69

enty

0.68

chool

0.63

oos

0.63

orts

0.63

alon

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.