INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ŃĶ

-0.83

enza

-0.78

risk

-0.74

ollower

-0.74

ELY

-0.73

¶ħ

-0.71

ributes

-0.71

plate

-0.70

roxy

-0.69

compl

-0.68

POSITIVE LOGITS

Mun

0.76

urrent

0.71

 Maker

0.66

ABE

0.62

 anim

0.61

Gat

0.60

TBA

0.60

ub

0.60

jun

0.60

Kee

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.