INDEX

Explanations

instances where someone or something is responsible for a specific action or outcome

oai_token-act-pair · gpt-3.5-turbo

phrases indicating accountability or responsibility

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

edin

-0.85

Sport

-0.81

ellen

-0.78

notations

-0.75

opol

-0.74

alk

-0.73

lime

-0.73

nets

-0.72

fair

-0.72

Brow

-0.70

POSITIVE LOGITS

 overseeing

1.04

 maintaining

1.01

 safegu

0.99

 regulating

0.99

 ensuring

0.96

 protecting

0.95

 preserving

0.90

 upholding

0.90

 constructing

0.89

 upkeep

0.88

Activations Density 0.069%

instances where someone or something is responsible for a specific action or outcome

phrases indicating accountability or responsibility

No Comments

No Known Activations