INDEX

Explanations

phrases expressing hope that something helps

oai_token-act-pair · gemini-2.0-flash

words expressing hope or assistance

np_token-act-pair-logits · gpt-4o-mini

foreign words

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

promise

-0.64

helpful

-0.63

 dyna

-0.61

 Alvarado

-0.61

EROS

-0.61

 promise

-0.60

 Promise

-0.60

unggu

-0.59

 resourceCulture

-0.59

 reversal

-0.59

POSITIVE LOGITS

 nationaux

0.57

providedIn

0.56

 might

0.55

 igång

0.54

توضیحات

0.54

 maybe

0.51

+#+#

0.50

may

0.50

 quizás

0.50

 militaires

0.50

Activations Density 1.417%

phrases expressing hope that something helps

words expressing hope or assistance

foreign words

No Comments

No Known Activations