INDEX

Explanations

phrases related to social dynamics and hierarchy, specifically focusing on deference and status

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

 charité

-0.64

 Settembre

-0.59

 akut

-0.56

 gmbh

-0.56

 Luglio

-0.54

 optik

-0.54

 errone

-0.54

PerformLayout

-0.53

 Queste

-0.53

 bonté

-0.53

POSITIVE LOGITS

 intentions

0.55

 embodi

0.54

ophosph

0.52

 signify

0.52

 intent

0.52

 wearer

0.50

 belonging

0.50

 seriousness

0.49

 MediatR

0.47

 intention

0.47

Activations Density 0.426%

phrases related to social dynamics and hierarchy, specifically focusing on deference and status

No Comments

No Known Activations