INDEX

Explanations

elements related to decision-making and consequence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

تقاوى

-0.59

 typelib

-0.59

webElementGuid

-0.58

saraba

-0.58

epam

-0.57

ModelAdmin

-0.56

/*

-0.55

unately

-0.55

__((

-0.54

ardless

-0.54

POSITIVE LOGITS

而是

1.23

 بلکه

1.21

 sondern

1.15

 Instead

0.91

 melainkan

0.91

 Rather

0.85

Instead

0.83

 instead

0.83

 אלא

0.81

むしろ

0.81

Activations Density 0.317%

elements related to decision-making and consequence

No Comments

No Known Activations