INDEX

Explanations

references to the term "light" in 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 @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

 propOrder

-1.20

 оригіналу

-1.03

AndEndTag

-0.97

ganu

-0.97

 bezeichneter

-0.96

enumii

-0.95

 Anfitrión

-0.95

 تانيه

-0.93

 blessés

-0.92

addPreferredGap

-0.91

POSITIVE LOGITS

 Light

1.57

 LIGHT

1.48

Light

1.48

light

1.44

 light

1.39

LIGHT

1.35

 Lightfoot

1.14

lights

1.09

 Lights

1.08

 Licht

1.07

Activations Density 0.061%

references to the term "light" in various contexts

No Comments

No Known Activations