INDEX

Explanations

references to "Je" or "je," indicating a focus on personal pronouns in the context of self-reference

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

NameInMap

-1.06

 समीक्षाओं

-1.04

uxxxx

-0.97

 insuffisamment

-0.92

Хьажоргаш

-0.92

UnusedPrivate

-0.91

Personendaten

-0.88

 تضيفلها

-0.88

 >=",

-0.83

发表于

-0.81

POSITIVE LOGITS

Jez

0.57

 suis

0.56

Je

0.51

ộ

0.50

Gom

0.47

Je

0.47

je

0.41

хова

0.39

JEE

0.38

 sais

0.38

Activations Density 0.085%

references to "Je" or "je," indicating a focus on personal pronouns in the context of self-reference

No Comments

No Known Activations