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

iris

-0.76

aria

-0.70

lein

-0.69

 istg

-0.66

unin

-0.65

 è£ıè¦ļéĨĴ

-0.65

#$#$

-0.65

 Dear

-0.65

vertis

-0.64

RAM

-0.64

POSITIVE LOGITS

hedral

0.82

 charism

0.78

.–

0.73

enary

0.69

 srfAttach

0.68

CLOSE

0.67

uling

0.66

olly

0.65

ista

0.65

 invention

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.