INDEX

Explanations

locations and states, specifically highlighting "Kansas" with high activations

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

 alho

-0.56

putes

-0.54

valente

-0.51

 Catawiki

-0.50

pegno

-0.49

ferrer

-0.48

assertIn

-0.48

ytick

-0.47

Thiết

-0.47

ceğim

-0.47

POSITIVE LOGITS

 Kansas

1.43

Kansas

1.32

 KANSAS

1.26

 Kans

1.01

 Missouri

0.96

KC

0.96

kc

0.90

 Topeka

0.90

Missouri

0.87

KU

0.78

Activations Density 0.146%

locations and states, specifically highlighting "Kansas" with high activations

No Comments

No Known Activations