INDEX

Explanations

instances of people speaking and being quoted

oai_token-act-pair · gemini-2.0-flash

quotations

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 typelib

-0.68

TemporalType

-0.67

talks

-0.64

basicConfig

-0.63

bní

-0.63

enumi

-0.61

sizeCache

-0.60

렷

-0.60

 talks

-0.60

 talk

-0.59

POSITIVE LOGITS

RTDA

0.48

écart

0.43

 considérée

0.42

のである

0.42

่ม

0.41

interno

0.41

ెక్

0.40

кен

0.40

 declared

0.40

 considéré

0.40

Activations Density 0.355%

instances of people speaking and being quoted

quotations

No Comments

No Known Activations