INDEX

Explanations

programming constructs related to method invocation and variable manipulation

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

⟬

-0.64

MathML

-0.59

enumi

-0.58

 unknownFields

-0.51

getResultList

-0.51

forName

-0.50

InstanceState

-0.50

rubin

-0.49

MultipartFile

-0.49

 aisladas

-0.49

POSITIVE LOGITS

IMPORTED

0.62

 ModelExpression

0.60

jsdelivr

0.60

 terça

0.59

UseVisualStyle

0.57

 insuffisamment

0.54

glTexCoord

0.53

iffance

0.52

AndroidJUnit

0.51

LookAnd

0.50

Activations Density 0.018%

programming constructs related to method invocation and variable manipulation

No Comments

No Known Activations