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

aez

-0.74

 Shap

-0.72

 pulp

-0.72

 ç¥ŀ

-0.68

PG

-0.67

 reprint

-0.66

 Monthly

-0.63

ppo

-0.62

 allowance

-0.61

SIL

-0.61

POSITIVE LOGITS

wagen

0.69

bred

0.69

isite

0.67

 Razor

0.67

 Heisman

0.65

fleet

0.64

conference

0.64

leading

0.64

halla

0.63

gravity

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.