INDEX

Explanations

words related to being suitable, specifically variations of the word "suitable."

oai_token-act-pair · gpt-3.5-turbo

words related to traits or characteristics, particularly in a quantitative or evaluative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

raped

-0.83

opsis

-0.82

romy

-0.79

trak

-0.78

aptic

-0.77

href

-0.73

utra

-0.72

het

-0.72

omet

-0.70

actionDate

-0.70

POSITIVE LOGITS

itable

1.30

itably

0.83

itability

0.71

 charism

0.71

 contribution

0.71

 Creature

0.70

 Haram

0.68

 trait

0.68

 Collider

0.68

 heterogeneity

0.68

Activations Density 0.009%

words related to being suitable, specifically variations of the word "suitable."

words related to traits or characteristics, particularly in a quantitative or evaluative context

No Comments

No Known Activations