INDEX

Explanations

repeated phrases indicating information that has been communicated or relayed

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

makeText

-0.61

addGroup

-0.52

 ويكيميديا

-0.52

 retain

-0.51

chaus

-0.51

QRect

-0.51

cee

-0.49

see

-0.48

 Nadu

-0.47

ivar

-0.47

POSITIVE LOGITS

 heard

3.52

heard

2.88

 Heard

2.81

Heard

2.76

 overheard

1.78

 escuchado

1.58

 listened

1.51

 hears

1.37

 entendu

1.36

 слыша

1.28

Activations Density 0.045%

repeated phrases indicating information that has been communicated or relayed

No Comments

No Known Activations