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

 Lights

-0.70

aired

-0.65

 Closed

-0.64

 Hollow

-0.63

 bugs

-0.60

Lam

-0.59

EFF

-0.58

anmar

-0.57

hen

-0.57

Tet

-0.57

POSITIVE LOGITS

uti

0.87

etheless

0.81

phabet

0.71

pass

0.71

zilla

0.69

inational

0.66

=~

0.65

hm

0.63

nine

0.62

theme

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.