INDEX

Explanations

phrases related to proving oneself or others in a competitive or challenging context

oai_token-act-pair · gpt-3.5-turbo

phrases related to self-proving and demonstrating competence

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

 IMAGES

-0.76

 heels

-0.68

inately

-0.68

anon

-0.63

catentry

-0.60

 Commonwealth

-0.60

 Thoughts

-0.60

NCT

-0.58

 shares

-0.58

 Lynn

-0.58

POSITIVE LOGITS

worthiness

0.76

Applic

0.70

cellence

0.68

classic

0.66

IER

0.65

 viability

0.65

 faithful

0.64

Fail

0.64

Ã©n

0.63

vantage

0.63

Activations Density 0.114%

phrases related to proving oneself or others in a competitive or challenging context

phrases related to self-proving and demonstrating competence

No Comments

No Known Activations