INDEX

Explanations

distinct grammatical markers or symbols indicative of comparisons or contrasts

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

är än

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

.*")]

-0.58

#+#

-0.56

cupertino

-0.54

-------------</

-0.49

 المعيارى

-0.48

ImageField

-0.46

yntaxException

-0.46

 informée

-0.46

uxxxx

-0.45

 aDecoder

-0.44

POSITIVE LOGITS

 &___

0.39

processable

0.39

 насељу

0.38

บาย

0.37

 parachoque

0.36

thâu

0.35

saraba

0.34

Nationalité

0.34

\[

0.34

における

0.34

Activations Density 0.000%

distinct grammatical markers or symbols indicative of comparisons or contrasts

är än

No Comments

No Known Activations