INDEX

Explanations

phrases related to physiological processes or bodily functions

oai_token-act-pair · gpt-3.5-turbo

elements of a cause-and-effect relationship in scientific or observational contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

abouts

-0.86

©¶æ

-0.74

izen

-0.70

spection

-0.69

raph

-0.65

chal

-0.64

quished

-0.63

ominated

-0.61

quartered

-0.60

usp

-0.59

POSITIVE LOGITS

 causing

1.52

 allowing

1.31

 thereby

1.30

 forcing

1.27

 resulting

1.25

 creating

1.25

 preventing

1.25

 exacerb

1.23

 enabling

1.22

 reducing

1.20

Activations Density 0.311%

phrases related to physiological processes or bodily functions

elements of a cause-and-effect relationship in scientific or observational contexts

No Comments

No Known Activations