INDEX

Explanations

expressions of anger and frustration

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

 otomatig

-0.81

 snippetHide

-0.75

findpost

-0.71

BeginContext

-0.71

 nahilalakip

-0.64

хьтан

-0.63

contentLoaded

-0.63

 intptr

-0.61

NameInMap

-0.60

 ModelExpression

-0.59

POSITIVE LOGITS

Angry

0.78

 angry

0.77

 anger

0.77

 Angry

0.75

 Anger

0.72

 colère

0.71

angry

0.70

 enoj

0.69

Anger

0.68

怒

0.65

Activations Density 0.438%

expressions of anger and frustration

No Known Activations