INDEX

Explanations

conversational language and expressions of opinion

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

AddTagHelper

-0.67

AnchorTagHelper

-0.67

 resourceCulture

-0.63

iſchen

-0.62

RenderAtEndOf

-0.62

niſſe

-0.62

iſche

-0.61

majánló

-0.60

 increí

-0.60

OrBuilder

-0.60

POSITIVE LOGITS

 kind

0.39

 sort

0.37

 Buchen

0.34

下

0.32

 entre

0.32

 مقاط

0.30

……

0.30

 catalog

0.30

 segu

0.30

...

0.30

Activations Density 0.137%

conversational language and expressions of opinion

No Comments

No Known Activations