INDEX

Explanations

specific technical and programming-related terms and concepts

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

featureID

-0.73

ConstraintMaker

-0.70

IntoConstraints

-0.67

OCCURRED

-0.67

BagConstraints

-0.62

 iNdEx

-0.60

цездатний

-0.59

новниш

-0.59

ftagPool

-0.59

UnsafeEnabled

-0.58

POSITIVE LOGITS

 dieux

0.39

 effek

0.38

 vérit

0.37

Editorial

0.35

AddUser

0.34

METHODOLOGY

0.33

 Editorial

0.33

 эффек

0.32

 chrétiens

0.32

 Lolita

0.32

Activations Density 0.001%

specific technical and programming-related terms and concepts

No Comments

No Known Activations