INDEX

Explanations

negative statements with emphasis on denial

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

Iš

-0.67

Į

-0.63

 Bár

-0.59

 Municipio

-0.59

 CommonModule

-0.58

 optik

-0.57

WebElementEntity

-0.57

Võ

-0.56

Ngb

-0.55

 Cár

-0.55

POSITIVE LOGITS

 vodi

0.66

 letti

0.64

 pensieri

0.61

 affez

0.60

 koš

0.60

 najbol

0.59

 nemici

0.59

 vestiti

0.59

 unlaw

0.59

 zima

0.59

Activations Density 0.125%

negative statements with emphasis on denial

No Comments

No Known Activations