INDEX

Explanations

phrases related to politics and government, including names of political figures, policy proposals, and official statements

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 malheure

-1.02

 étrang

-1.00

 carrefour

-1.00

 plong

-0.97

 malheureux

-0.93

 prétend

-0.88

 héro

-0.86

 cahier

-0.84

hcm

-0.83

 Miscell

-0.82

POSITIVE LOGITS

 aware

0.73

 able

0.72

 glad

0.71

 ready

0.71

 willing

0.71

 afraid

0.68

 gonna

0.68

 proud

0.68

 unable

0.67

 pleased

0.67

Activations Density 0.249%

phrases related to politics and government, including names of political figures, policy proposals, and official statements

No Comments

No Known Activations