INDEX

Explanations

phrases that refer to specific groups of people or medical conditions

oai_token-act-pair · gpt-3.5-turbo

references to individuals or groups with various attributes or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

orpor

-0.66

 promul

-0.64

pher

-0.61

 Authors

-0.60

among

-0.59

wark

-0.59

ecause

-0.58

finals

-0.58

oner

-0.57

once

-0.57

POSITIVE LOGITS

 disabilities

1.19

 propensity

1.01

 penchant

0.94

 disability

0.94

 vested

0.92

 backgrounds

0.91

 allergies

0.90

 predis

0.90

 knack

0.88

 incomes

0.88

Activations Density 0.219%

phrases that refer to specific groups of people or medical conditions

references to individuals or groups with various attributes or conditions

No Comments

No Known Activations