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

SourceFile

-0.70

itech

-0.67

 overlook

-0.65

ãĥ¼ãĤ¯

-0.64

Plugin

-0.63

à©

-0.63

/>

-0.62

Hidden

-0.62

 Neck

-0.62

 synth

-0.62

POSITIVE LOGITS

ascus

0.72

imposed

0.71

amaz

0.66

enium

0.64

raved

0.63

iddled

0.62

STE

0.62

raint

0.62

 #####

0.61

riter

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.