INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 GOODMAN

-0.68

ONES

-0.63

 Swim

-0.61

med

-0.61

 Transgender

-0.60

olls

-0.60

roleum

-0.59

 pharm

-0.59

 Salary

-0.58

 biom

-0.58

POSITIVE LOGITS

nit

0.83

cel

0.82

rev

0.79

Fed

0.78

maxwell

0.76

luster

0.73

gur

0.73

Russ

0.72

Fl

0.72

xit

0.67

Activations Density 0.000%

No Known Activations

This feature has no known activations.