INDEX

Explanations

words related to negative judgments or criticisms

oai_token-act-pair · gpt-3.5-turbo

terms related to disgrace and moral failure

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 ellipt

-0.89

 envelope

-0.77

 Avalon

-0.72

 ultrasound

-0.65

apt

-0.63

 equilibrium

-0.62

 impulse

-0.62

 airplane

-0.62

addafi

-0.61

 shutter

-0.61

POSITIVE LOGITS

ful

1.41

fully

1.15

orial

0.94

ious

0.94

beat

0.89

xual

0.87

yers

0.84

edly

0.83

forth

0.83

es

0.82

Activations Density 0.045%

words related to negative judgments or criticisms

terms related to disgrace and moral failure

No Comments

No Known Activations