INDEX

Explanations

sentences where new information or evidence is given

oai_token-act-pair · gemini-2.0-flash

technical details related to computer code

np_token-act-pair-logits · gemini-2.0-flash

giving information

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_16/width_16k/average_l0_10

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

 EClass

-0.64

StoryboardSegue

-0.59

ામ

-0.49

ItemBackground

-0.49

jeeling

-0.45

kubectl

-0.45

ioare

-0.44

 indetermin

-0.42

uern

-0.42

 given

-0.42

POSITIVE LOGITS

Personendaten

0.63

limus

0.57

middleware

0.56

esModule

0.56

 sàng

0.56

 Audiodateien

0.54

cèse

0.54

 unknownFields

0.53

WithIdentifier

0.52

 connexe

0.51

Activations Density 1.244%

sentences where new information or evidence is given

technical details related to computer code

giving information

No Comments

No Known Activations