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

interstitial

-0.76

 obser

-0.73

boa

-0.73

 Myster

-0.72

undo

-0.72

 Telegram

-0.67

 Benefit

-0.66

ographics

-0.65

iasis

-0.64

ographic

-0.63

POSITIVE LOGITS

Poc

0.73

MJ

0.71

Osw

0.70

HL

0.70

 awakening

0.64

Occup

0.63

 pacif

0.63

 euth

0.63

CF

0.61

aukee

0.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.