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

-0.88

FORE

-0.66

[&

-0.65

iland

-0.64

SHARE

-0.64

 clos

-0.63

byss

-0.62

ById

-0.61

Privacy

-0.61

Reloaded

-0.61

POSITIVE LOGITS

ordan

0.64

ulton

0.59

olesc

0.59

 Ethics

0.58

esses

0.58

 audition

0.57

lim

0.57

lime

0.57

 enrolled

0.57

 Mandarin

0.56

Activations Density 0.000%

No Known Activations

This feature has no known activations.