INDEX

Explanations

symbols or markers that indicate specific data structures or functions in programming

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

hex values and code constants

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

⤹

-0.46

īdz

-0.44

msgSender

-0.43

MockBean

-0.43

ocl

-0.40

ództ

-0.39

 propOrder

-0.39

zicht

-0.39

fsen

-0.39

lectricité

-0.38

POSITIVE LOGITS

AccessorTable

0.64

tagHelperRunner

0.55

complexContent

0.52

enterOuterAlt

0.52

 kasarigan

0.52

 كومونز

0.50

Παραπομπές

0.50

Legături

0.48

 CURLOPT

0.48

IntoConstraints

0.48

Activations Density 0.006%

symbols or markers that indicate specific data structures or functions in programming

hex values and code constants

No Comments

No Known Activations