INDEX

Explanations

terms related to military or geographical categorizations and complex scientific or technical explanations

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

 <<<<<<<<<<<<<<

-0.73

hasRole

-0.60

 GenerationType

-0.55

rceil

-0.55

 Picchu

-0.54

bitField

-0.54

TypeDef

-0.54

prefixer

-0.54

mnar

-0.54

ORPG

-0.53

POSITIVE LOGITS

 raiſ

0.68

 eksp

0.67

stawa

0.65

EXP

0.63

گزاری

0.61

 reafon

0.60

exp

0.60

 sjø

0.59

postmedia

0.59

 EXPL

0.59

Activations Density 0.091%

terms related to military or geographical categorizations and complex scientific or technical explanations

No Comments

No Known Activations