INDEX

Explanations

the word "do", sometimes in combination with "server", "be", or "think"

oai_token-act-pair · gemini-2.0-flash

do

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

RenderAtEndOf

-1.05

ymce

-0.85

atecas

-0.82

 mergeFrom

-0.81

Vidite

-0.81

 disambiguazione

-0.78

<bos>

-0.76

 jScrollPane

-0.74

RectangleBorder

-0.71

 JpaRepository

-0.71

POSITIVE LOGITS

 does

0.72

does

0.65

 eût

0.62

 paraître

0.62

 Does

0.62

 auroit

0.60

did

0.59

 mourut

0.59

did

0.59

Does

0.59

Activations Density 0.325%

the word "do", sometimes in combination with "server", "be", or "think"

do

No Comments

No Known Activations