INDEX

Explanations

phrases or words that indicate repetition or recurring actions

oai_token-act-pair · gpt-3.5-turbo

the repetition of phrases involving "over and over again."

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

Negative Logits

VICE

-0.68

icho

-0.63

antz

-0.60

asus

-0.59

mosp

-0.58

ropri

-0.57

lyn

-0.57

 pestic

-0.56

 Medic

-0.56

riott

-0.56

POSITIVE LOGITS

 again

1.43

again

1.14

AGA

1.04

 Again

0.98

 until

0.91

 anew

0.90

etheless

0.84

 thereafter

0.82

Again

0.80

 till

0.78

Activations Density 0.066%

phrases or words that indicate repetition or recurring actions

the repetition of phrases involving "over and over again."

No Known Activations