INDEX

Explanations

references to conflict, animosity, and rivalry between individuals or groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 snippetHide

-0.49

цездатний

-0.49

 computadoras

-0.48

 autorytatywna

-0.48

醐

-0.46

posedge

-0.46

 nonUne

-0.46

denn

-0.46

DoubleQuotes

-0.45

Географиясе

-0.44

POSITIVE LOGITS

 hatred

0.56

 hate

0.54

 odio

0.54

 hates

0.50

hate

0.49

 hating

0.47

 hated

0.46

 Hate

0.45

 haine

0.44

 HATE

0.42

Activations Density 0.402%

references to conflict, animosity, and rivalry between individuals or groups

No Comments

No Known Activations