INDEX

Explanations

instances of the word "notes" in sentences

oai_token-act-pair · gpt-3.5-turbo

repeated references to "notes" in the text

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

LOS

-0.86

nom

-0.65

 volunte

-0.64

 multif

-0.64

Interstitial

-0.60

 judicial

-0.59

nex

-0.59

 displ

-0.58

 outp

-0.57

shr

-0.57

POSITIVE LOGITS

 notes

1.24

notes

1.16

 Notes

1.10

 note

0.89

books

0.88

Notes

0.87

itaire

0.79

iewicz

0.77

manship

0.74

note

0.74

Activations Density 0.011%

instances of the word "notes" in sentences

repeated references to "notes" in the text

No Comments

No Known Activations