INDEX

Explanations

structured data elements, possibly related to file handling or data processing functions

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

Code snippets with brackets or file paths

np_acts-logits-general · gemini-2.0-flash

calibration measurement

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

PYX

-0.61

 مرئيه

-0.51

GraphicsUnit

-0.48

 getopt

-0.46

 informée

-0.46

Collar

-0.44

enumi

-0.43

 habido

-0.43

tagHelperRunner

-0.43

engaruhi

-0.42

POSITIVE LOGITS

WriteTagHelper

0.67

PMailer

0.59

æn

0.58

styleType

0.58

 мәкал

0.57

polation

0.56

cembre

0.56

RenderAtEndOf

0.55

örande

0.54

JvmStatic

0.54

Activations Density 0.042%

structured data elements, possibly related to file handling or data processing functions

Code snippets with brackets or file paths

calibration measurement

No Comments

No Known Activations