INDEX

Explanations

phrases related to political and social commentary, including terms like "cultural critique," "religious and racial categories," and "immigration sentiments."

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

 bordeaux

-1.46

Juf

-1.43

 fluo

-1.41

 franz

-1.35

 casio

-1.34

 lyon

-1.32

 dises

-1.30

 Châ

-1.30

 canel

-1.29

 levis

-1.29

POSITIVE LOGITS

 doesn

0.85

 happens

0.80

 seems

0.80

 helps

0.79

It

0.79

 enables

0.79

is

0.79

 allows

0.79

 does

0.78

 represents

0.78

Activations Density 0.232%

phrases related to political and social commentary, including terms like "cultural critique," "religious and racial categories," and "immigration sentiments."

No Comments

No Known Activations