INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ickets

-0.72

 PRESS

-0.65

 contam

-0.64

emort

-0.64

anie

-0.64

idth

-0.63

hus

-0.60

oku

-0.60

{*

-0.60

IFF

-0.59

POSITIVE LOGITS

Lago

0.75

ancial

0.68

 Nanto

0.66

 envy

0.63

stanbul

0.62

cue

0.60

 Fiat

0.60

 cellar

0.59

 Canaver

0.59

nell

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.