INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Mald

-0.65

 coerc

-0.61

 Pagan

-0.60

 Investig

-0.60

NULL

-0.59

 unsub

-0.58

dom

-0.58

 appellant

-0.58

vertisements

-0.57

ministic

-0.56

POSITIVE LOGITS

 Flavoring

0.93

ocular

0.81

cule

0.76

ozy

0.70

Leaks

0.70

catentry

0.68

Jr

0.67

speech

0.67

utters

0.66

oot

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.