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

imens

-0.85

phas

-0.76

egu

-0.76

act

-0.73

inct

-0.70

avi

-0.69

 Massacre

-0.67

 Synd

-0.67

iments

-0.66

eng

-0.65

POSITIVE LOGITS

 Sims

0.73

uler

0.71

 oven

0.71

TRY

0.66

 furnace

0.64

 paycheck

0.64

 Solomon

0.63

 dilig

0.63

 frying

0.63

 diction

0.62

Activations Density 0.000%

No Known Activations

This feature has no known activations.