INDEX

Explanations

references to the word "Red" and its various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

ConstraintMaker

-1.05

makeConstraints

-0.81

Autoritní

-0.75

elemField

-0.71

 AssemblyCompany

-0.67

RectangleBorder

-0.67

Enllaces

-0.67

 jScrollPane

-0.65

AccessorTable

-0.62

 VIAF

-0.62

POSITIVE LOGITS

RED

1.00

Red

0.95

Red

0.90

 レッド

0.78

RED

0.77

 rosse

0.76

red

0.74

 erythro

0.71

 Reddy

0.69

 vermelha

0.67

Activations Density 0.080%

references to the word "Red" and its various contexts

No Comments

No Known Activations