INDEX

Explanations

references to online resources and contact information related to a user or organization

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

Following "the" in code or websites

np_acts-logits-general · gemini-2.0-flash

appropriateness of sections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

WireFormat

-0.65

 مشين

-0.65

 bParam

-0.64

 CURIAM

-0.60

IntoConstraints

-0.60

 uintptr

-0.60

hatenablog

-0.59

expandindo

-0.58

rangian

-0.56

writeFieldEnd

-0.56

POSITIVE LOGITS

 appropriate

0.99

appropriate

0.76

 respective

0.73

 relevant

0.69

相应

0.65

 Appropriate

0.62

“

0.61

 appropri

0.61

 соответству

0.60

相应的

0.60

Activations Density 0.433%

references to online resources and contact information related to a user or organization

Following "the" in code or websites

appropriateness of sections

No Comments

No Known Activations