INDEX

Explanations

phrases related to hate and negativity

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

 المعيارى

-0.58

Composição

-0.53

FieldNumber

-0.50

ResumeLayout

-0.49

tisgarh

-0.49

IntegerField

-0.48

CharField

-0.48

Apesar

-0.47

Horário

-0.47

ApiProperty

-0.47

POSITIVE LOGITS

 maneu

1.20

 depic

1.13

 curé

1.13

 unve

1.13

 HATE

1.13

 encomp

1.12

 Abbé

1.12

 hairc

1.10

 affor

1.09

 increa

1.09

Activations Density 0.104%

phrases related to hate and negativity

No Comments

No Known Activations