INDEX

Explanations

warning signs of potential harm or violence, especially in the context of social isolation and manipulation

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

 libéral

-0.55

erenc

-0.54

ladiator

-0.54

 souverain

-0.52

 décret

-0.51

 initComponents

-0.51

 Portail

-0.50

 dimenti

-0.49

Tembelea

-0.49

klere

-0.48

POSITIVE LOGITS

 suspicious

0.87

 suspicion

0.72

 suspect

0.72

 suspicions

0.68

 suspected

0.64

 sospe

0.63

 arouse

0.62

 unusual

0.61

 anomalous

0.58

 alerted

0.57

Activations Density 0.672%

warning signs of potential harm or violence, especially in the context of social isolation and manipulation

No Comments

No Known Activations