INDEX

Explanations

references to responsibilities and duties within various contexts

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

himovic

-0.48

piram

-0.46

 sonrisa

-0.44

grimas

-0.42

asaran

-0.40

 segar

-0.40

 vainqueur

-0.39

__*/

-0.39

 hyped

-0.39

unmodifiable

-0.39

POSITIVE LOGITS

 duties

1.19

 duty

1.10

 tasks

1.01

 Duties

0.98

 tugas

0.92

duty

0.91

 task

0.90

Duties

0.90

 respon

0.89

 role

0.88

Activations Density 0.458%

references to responsibilities and duties within various contexts

No Comments

No Known Activations