INDEX

Explanations

phrases indicating likelihood or probability

oai_token-act-pair · gpt-3.5-turbo

phrases related to likelihood or probability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

æ©

-0.86

idelines

-0.82

æ©Ł

-0.81

agall

-0.79

çļ

-0.78

NCT

-0.76

Introduced

-0.72

âĸ¬

-0.70

Tar

-0.68

ļéĨĴ

-0.68

POSITIVE LOGITS

 future

0.81

 someday

0.81

 Paradise

0.78

 accidental

0.69

 wiser

0.68

 outcome

0.68

 runaway

0.66

 unlucky

0.66

 forgiven

0.66

 impending

0.65

Activations Density 0.172%

phrases indicating likelihood or probability

phrases related to likelihood or probability

No Comments

No Known Activations