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

etheless

-0.83

ilogy

-0.76

 veter

-0.76

 suspic

-0.74

 nodd

-0.74

 metic

-0.70

esta

-0.69

erning

-0.69

achus

-0.68

 perpend

-0.68

POSITIVE LOGITS

imental

0.71

ragon

0.64

)[

0.63

odon

0.61

DOI

0.61

 Grateful

0.61

ange

0.60

Æ

0.60

eq

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.