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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

pse

-0.83

 sophistic

-0.79

mas

-0.77

 imperson

-0.77

 absor

-0.77

 fortun

-0.76

ector

-0.76

mir

-0.73

 bureaucr

-0.72

 credential

-0.71

POSITIVE LOGITS

ilk

0.86

ilities

0.83

ournals

0.78

 Milk

0.75

illation

0.73

uca

0.73

ptoms

0.73

ELL

0.72

 Kaufman

0.71

poons

0.70

Activations Density 0.000%

No Known Activations

This feature has no known activations.