INDEX

Explanations

terms and phrases related to legal procedures and actions

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

or listing alternatives

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

PositiveButton

-0.54

ApiModelProperty

-0.49

 nød

-0.49

assertRaises

-0.48

 оригіналу

-0.48

RotationOrder

-0.47

verifyException

-0.47

 Trip

-0.46

 rhestr

-0.44

 Himo

-0.44

POSITIVE LOGITS

styleType

0.52

 intStringLen

0.43

ChildScrollView

0.41

yntaxException

0.40

UALA

0.35

 Sebab

0.35

setVerticalGroup

0.35

VYMaps

0.34

 ControllerBase

0.34

 Wiktionnaire

0.33

Activations Density 0.114%

terms and phrases related to legal procedures and actions

or listing alternatives

No Comments

No Known Activations