INDEX

Explanations

the beginning and end of text blocks or sections

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 unspeak

-2.62

 reluct

-2.41

 unlaw

-2.40

 disgra

-2.40

 impractica

-2.28

 shenan

-2.28

 ineffec

-2.26

 horrend

-2.25

 impra

-2.21

 belliger

-2.18

POSITIVE LOGITS

<bos>

14.43

GEBURTSDATUM

2.54

expandindo

2.53

 betweenstory

2.48

Autoritní

2.46

تقاوى

2.20

 Administrativna

2.16

 Italijani

2.15

 kasarigan

2.11

 Paglinawan

2.11

Activations Density 0.090%

the beginning and end of text blocks or sections

No Comments

No Known Activations