INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

igree

-0.73

minded

-0.70

esome

-0.70

ulas

-0.65

pants

-0.64

 paternal

-0.62

rella

-0.62

 pins

-0.62

 minded

-0.61

theme

-0.61

POSITIVE LOGITS

 âĸº

0.73

 Analy

0.72

 Deter

0.70

 Probe

0.69

 Simulation

0.66

 Detention

0.65

Lear

0.64

 Guilty

0.64

Grab

0.64

EVA

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.