INDEX

Explanations

negative and contrasting statements

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

 Ibidem

-0.51

 Simult

-0.49

BeforeMethod

-0.47

Vys

-0.47

AppCompatTheme

-0.46

CompoundButton

-0.46

InstrumentedTest

-0.44

 Hän

-0.44

 Opis

-0.44

 Enlarged

-0.43

POSITIVE LOGITS

 necessarily

0.57

 necesar

0.54

 autorytatywna

0.49

but

0.48

rewsbury

0.48

 jajaja

0.47

但不

0.47

ghz

0.46

ongevity

0.46

 burberry

0.46

Activations Density 0.130%

negative and contrasting statements

No Comments

No Known Activations