INDEX

Explanations

expressions of frustration or anger in dialogue

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 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

 IndexPath

-0.66

ValueStyle

-0.63

 informée

-0.59

 varandra

-0.56

 orejas

-0.56

WithIOException

-0.55

 masculinos

-0.55

 cejas

-0.54

kijken

-0.54

 snippetHide

-0.54

POSITIVE LOGITS

NO

0.39

 words

0.36

 shouting

0.35

bak

0.34

 wrong

0.33

 scolded

0.33

Dammit

0.33

 shouted

0.32

 cries

0.32

 shouts

0.32

Activations Density 0.170%

expressions of frustration or anger in dialogue

No Comments

No Known Activations