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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Reference

-0.81

ially

-0.76

 Profile

-0.76

cyclopedia

-0.75

ogy

-0.75

agi

-0.71

olia

-0.69

odium

-0.67

bsite

-0.67

iated

-0.66

POSITIVE LOGITS

 Thom

0.78

 Medals

0.74

*/(

0.68

 Panthers

0.66

 Mitch

0.64

Â¯Â¯Â¯Â¯

0.63

 Blitz

0.61

 Dante

0.60

 Punk

0.59

 Petty

0.59

Activations Density 0.000%

No Known Activations

This feature has no known activations.