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

abil

-0.84

MX

-0.77

udos

-0.70

 Giul

-0.70

 Assist

-0.69

erg

-0.67

eva

-0.67

 Beaut

-0.66

 undone

-0.65

-0.64

POSITIVE LOGITS

 majority

1.13

 glim

0.86

majority

0.84

 condem

0.69

 nomine

0.68

 commissioner

0.68

 parade

0.68

 curfew

0.68

 majorities

0.68

reau

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.