INDEX

Explanations

references to banking, financial terms, and account details

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

Code, identifiers, or XML snippets

np_acts-logits-general · gemini-2.0-flash

measure, memory, message, favorite

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

-0.38

ing

-0.34

as

-0.33

-0.32

try

-0.31

ul

-0.30

ese

-0.30

Ke

-0.30

ad

-0.30

POSITIVE LOGITS

AndEndTag

0.87

ंदीखरीदारी

0.85

 CreateTagHelper

0.84

 disambiguazione

0.77

 ujednoznacz

0.77

 ProtoMessage

0.76

########.

0.73

 JpaRepository

0.73

OGND

0.73

<unused43>

0.72

Activations Density 1.344%

references to banking, financial terms, and account details

Code, identifiers, or XML snippets

measure, memory, message, favorite

No Comments

No Known Activations