INDEX

Explanations

pronouns and verbs related to self-identity

oai_token-act-pair · gpt-3.5-turbo

references to personal identity and self-identification

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 temptation

-0.79

aptic

-0.66

 pitfalls

-0.64

 passage

-0.64

hiba

-0.63

 overdose

-0.63

 showers

-0.63

onga

-0.62

pmwiki

-0.62

 dosage

-0.62

POSITIVE LOGITS

pires

0.79

selves

0.77

 adherent

0.74

los

0.73

"@

0.72

"#

0.71

 subscrib

0.71

 "$:/

0.71

 unbeat

0.70

çīĪ

0.69

Activations Density 0.066%

pronouns and verbs related to self-identity

references to personal identity and self-identification

No Comments

No Known Activations