INDEX

Explanations

expressions of frustration and strong emotional reactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 35-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_35/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.35.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

неопр

-0.47

henswürdigkeiten

-0.46

 itſelf

-0.46

)•

-0.43

=@"

-0.41

 يتيمه

-0.41

 ſtill

-0.41

 myſelf

-0.41

Personendaten

-0.40

+")

-0.40

POSITIVE LOGITS

 damn

2.34

 fucking

2.19

 goddamn

2.17

 damned

2.09

 fuck

1.96

damn

1.95

 fuckin

1.92

 DAMN

1.88

 Damn

1.88

fucking

1.88

Activations Density 0.461%

expressions of frustration and strong emotional reactions

No Comments

No Known Activations