INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

AKING

-0.71

 Founders

-0.69

approved

-0.67

worldly

-0.65

OTA

-0.64

otle

-0.62

chy

-0.60

 deductible

-0.60

perture

-0.60

ighth

-0.59

POSITIVE LOGITS

schild

0.77

 Hunger

0.71

uberty

0.70

 guiActiveUn

0.69

ment

0.68

igation

0.66

 hunger

0.66

bom

0.65

gered

0.64

bern

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.