INDEX

Explanations

phrases related to understanding and knowledge accumulation

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

 labores

-0.37

现

-0.37

ksessa

-0.35

CAUTIONS

-0.35

Coarse

-0.34

 ProtoMessage

-0.33

LETIN

-0.33

병

-0.32

 later

-0.32

IsRequired

-0.32

POSITIVE LOGITS

ChildScrollView

0.69

 محفوظة

0.65

LookAnd

0.62

AccessorTable

0.55

aarrggbb

0.55

FieldBuilder

0.54

EDEFAULT

0.53

OGND

0.52

tagHelperRunner

0.52

AnimationsModule

0.52

Activations Density 0.031%

phrases related to understanding and knowledge accumulation

No Comments

No Known Activations