INDEX

Explanations

programming-related syntax and structure

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

XML tags and attributes

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 Paglinawan

-0.59

writeFieldEnd

-0.48

IntoConstraints

-0.47

PMailer

-0.47

tanleria

-0.46

URLException

-0.45

 himſelf

-0.45

GraphicsUnit

-0.45

 cooperación

-0.45

 coopération

-0.45

POSITIVE LOGITS

RuleContext

0.40

 customised

0.38

 centred

0.38

andinav

0.36

 Miners

0.36

ABET

0.36

 uska

0.35

mintaan

0.35

 Custom

0.34

 Slee

0.34

Activations Density 0.129%

programming-related syntax and structure

XML tags and attributes

No Comments

No Known Activations