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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

abases

-0.74

advertising

-0.72

\\\\\\\\

-0.70

thood

-0.67

ensible

-0.66

 reliance

-0.65

casts

-0.65

iciary

-0.65

ciplinary

-0.63

missions

-0.63

POSITIVE LOGITS

 scen

0.65

 Sass

0.64

 Parables

0.64

 STATES

0.64

 heartbeat

0.63

irlf

0.62

 Guth

0.62

 rainy

0.62

heimer

0.61

Miy

0.57

Activations Density 0.000%

No Known Activations

This feature has no known activations.