INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

spons

-0.77

ternity

-0.75

mercial

-0.73

ndum

-0.71

illance

-0.70

ufact

-0.70

til

-0.69

opausal

-0.67

prus

-0.67

 occas

-0.66

POSITIVE LOGITS

eson

0.69

ISTORY

0.66

appa

0.65

uay

0.63

 aliases

0.62

blood

0.61

 caric

0.59

eret

0.59

 personalities

0.59

ered

0.58

Activations Density 0.000%

No Known Activations

This feature has no known activations.