INDEX

Explanations

phrases or terms related to organizations or projects

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

Hentet

-0.82

 Wiktionnaire

-0.66

Preço

-0.62

Datuak

-0.59

 Савезне

-0.58

Hvad

-0.58

Географиясе

-0.57

bibnamefont

-0.57

Varför

-0.56

 متعلقه

-0.56

POSITIVE LOGITS

 guarante

1.06

 emphat

0.94

fuo

0.93

 fatis

0.91

 effe

0.89

 maneu

0.88

 Græ

0.88

 encomp

0.87

 inder

0.86

 attemp

0.85

Activations Density 0.292%

phrases or terms related to organizations or projects

No Comments

No Known Activations