INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

merce

-0.72

BCC

-0.69

 endif

-0.68

zyme

-0.67

uren

-0.66

MC

-0.65

Kau

-0.63

 Atlantis

-0.62

atur

-0.61

gments

-0.61

POSITIVE LOGITS

dylib

0.80

BRE

0.76

ascript

0.73

yip

0.73

luaj

0.71

otti

0.68

EStreamFrame

0.67

ivalry

0.67

 journalism

0.66

arist

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.