INDEX

Explanations

phrases indicating reasons or explanations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

endregion

-0.69

FormTagHelper

-0.58

méno

-0.56

 Junto

-0.56

 perpetuity

-0.56

Spoljašnje

-0.53

 fathoms

-0.53

stalt

-0.52

 endregion

-0.52

mappedBy

-0.52

POSITIVE LOGITS

 BoxDecoration

0.54

 work

0.53

ьаж

0.51

 utafitiHapana

0.50

spaceBetween

0.50

 strategy

0.47

 hart

0.47

strategy

0.46

coso

0.46

tus

0.46

Activations Density 0.345%

phrases indicating reasons or explanations

No Comments

No Known Activations