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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

eva

-0.71

EVA

-0.66

 Catalyst

-0.63

pr

-0.62

 Guer

-0.62

anyon

-0.60

rium

-0.60

 Liter

-0.59

 medi

-0.59

real

-0.58

POSITIVE LOGITS

ertodd

0.79

çİĭ

0.72

Ļ

0.69

é¾

0.67

ological

0.65

sburg

0.64

ŃĶ

0.63

terness

0.63

accompan

0.62

redd

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.