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

 fragmentation

-0.77

etheless

-0.76

ichick

-0.69

proof

-0.62

spawn

-0.60

atten

-0.60

hner

-0.59

 Lanka

-0.59

 traject

-0.58

hunt

-0.58

POSITIVE LOGITS

istration

0.70

usha

0.69

 cents

0.68

gt

0.66

va

0.64

ttle

0.61

ace

0.61

ÂłÂłÂłÂł

0.61

Business

0.60

acity

0.60

Activations Density 0.000%

No Known Activations

This feature has no known activations.