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

iary

-0.80

apse

-0.79

edly

-0.77

ournal

-0.73

ensable

-0.72

utterstock

-0.72

munition

-0.71

escent

-0.68

ystem

-0.67

osate

-0.66

POSITIVE LOGITS

alls

0.72

ELS

0.71

RY

0.70

all

0.69

WN

0.68

ALS

0.67

NetMessage

0.66

RAY

0.65

Bad

0.64

 highs

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.