INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 increments

-0.70

aps

-0.69

cess

-0.69

acs

-0.68

oles

-0.66

 smoot

-0.66

sets

-0.62

icular

-0.62

els

-0.61

ys

-0.60

POSITIVE LOGITS

SPA

0.82

indal

0.78

 VIDEOS

0.73

ortunately

0.70

anwhile

0.70

 conduc

0.69

 Codec

0.68

awa

0.67

 acknow

0.65

 Antar

0.64

Activations Density 0.000%

No Known Activations

This feature has no known activations.