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

rahim

-0.73

 guiName

-0.69

nir

-0.69

planes

-0.68

reck

-0.67

ineries

-0.66

ims

-0.65

aldi

-0.64

ernels

-0.64

latest

-0.63

POSITIVE LOGITS

 conversion

0.71

 arrang

0.68

\-

0.68

[|

0.63

Ops

0.63

 expel

0.62

Camp

0.62

 change

0.61

cation

0.61

 ward

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.