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

MJ

-0.66

 Joker

-0.62

CS

-0.61

peed

-0.60

DJ

-0.60

LCS

-0.60

REC

-0.59

PG

-0.59

ital

-0.59

 flare

-0.58

POSITIVE LOGITS

iatric

0.86

 locality

0.71

iosis

0.66

 ingred

0.65

likely

0.64

natureconservancy

0.64

soDeliveryDate

0.61

beh

0.60

ogh

0.60

iat

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.