INDEX

Explanations

the beginning of a paragraph or section in a document

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

-2.51

 unlaw

-2.29

 disgra

-2.29

 reluct

-2.27

 impractica

-2.17

 ineffec

-2.16

 horrend

-2.16

 shenan

-2.15

 belliger

-2.12

 impra

-2.08

POSITIVE LOGITS

<bos>

14.41

GEBURTSDATUM

2.54

expandindo

2.53

 betweenstory

2.48

Autoritní

2.45

تقاوى

2.19

 Administrativna

2.15

 Italijani

2.14

 kasarigan

2.11

 Paglinawan

2.09

Activations Density 0.088%

the beginning of a paragraph or section in a document

No Comments

No Known Activations