INDEX

Explanations

words related to an oversimplified or exaggerated representation of something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "overs" and its variations in contexts related to over-simplification or overdose concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

fighter

-0.72

venge

-0.72

lime

-0.71

Guard

-0.70

ility

-0.69

 Pirate

-0.69

 Vatican

-0.64

ulic

-0.64

wife

-0.63

 Grizz

-0.63

POSITIVE LOGITS

 overs

4.23

overs

2.02

 Overs

1.52

 overd

1.50

 overc

1.35

over

1.30

 underest

1.22

 overest

1.16

ove

1.15

 misunderstand

1.09

Activations Density 0.014%

words related to an oversimplified or exaggerated representation of something

instances of the word "overs" and its variations in contexts related to over-simplification or overdose concepts

No Comments

No Known Activations