INDEX

Explanations

directions or suggestions related to actions

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

suggestions and commands

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

UnitTesting

-0.52

pitaux

-0.49

RegressionTest

-0.42

IActionResult

-0.39

genodigd

-0.39

să

-0.37

readystatechange

-0.37

省市镇

-0.36

 snippetHide

-0.36

InstrumentedTest

-0.35

POSITIVE LOGITS

ंदीखरीदारी

0.54

 noqa

0.50

rawDesc

0.47

CreateModel

0.46

ISupport

0.45

tawesome

0.45

LabelTagHelper

0.44

帖最后由

0.44

RTSN

0.43

}{@

0.43

Activations Density 0.224%

directions or suggestions related to actions

suggestions and commands

No Comments

No Known Activations