INDEX

Explanations

references to individuals involved in actions or legal circumstances

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

 appels

-0.47

 tasting

-0.46

 insegna

-0.44

 joaat

-0.44

 fischer

-0.43

 appartenant

-0.42

require

-0.42

 partecipazione

-0.41

 Require

-0.41

 miteinander

-0.41

POSITIVE LOGITS

complexContent

0.91

enumi

0.89

tvguidetime

0.78

SourceChecksum

0.77

 deal

0.77

Демографія

0.73

 BrowserModule

0.71

有关

0.71

mybatisplus

0.71

 dealt

0.71

Activations Density 0.066%

references to individuals involved in actions or legal circumstances

No Comments

No Known Activations