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

 enqu

-0.74

ffield

-0.71

ovan

-0.69

orkshire

-0.68

arest

-0.66

 unsus

-0.66

icum

-0.65

ector

-0.65

iaries

-0.65

ically

-0.64

POSITIVE LOGITS

 UNHCR

0.74

Khe

0.68

ilon

0.68

Ĥª

0.68

 Emerson

0.67

ģĸ

0.67

 Mattis

0.66

Fight

0.64

 mitigating

0.64

 TOUR

0.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.