INDEX

Explanations

references to data privacy and compliance

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

actions or proceedings

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

 Administrativna

-1.13

Hochspringen

-0.58

 للمعارف

-0.56

GTCX

-0.55

 المعيارى

-0.54

 betweenstory

-0.50

 <>",

-0.50

Notae

-0.49

⅘

-0.47

makeConstraints

-0.47

POSITIVE LOGITS

tagHelperRunner

0.60

RenderAtEndOf

0.45

 pinulongan

0.43

 Forder

0.41

Abit

0.39

TagMode

0.37

ngrx

0.37

onViewCreated

0.37

andExpect

0.37

 dàng

0.37

Activations Density 0.078%

references to data privacy and compliance

actions or proceedings

No Comments

No Known Activations