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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

netflix

-0.84

 Genie

-0.74

aeda

-0.73

omb

-0.69

mega

-0.69

OPS

-0.68

ahime

-0.65

steen

-0.65

omatic

-0.65

ilda

-0.65

POSITIVE LOGITS

 reset

0.70

 init

0.65

Otherwise

0.65

®

0.63

College

0.62

Whenever

0.61

Update

0.61

Bind

0.60

Submit

0.59

 unrestricted

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.