INDEX

Explanations

phrases related to analysis and evaluation of processes or functions

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

following "phase" or "given"

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

oredCriteria

-0.52

complexContent

-0.48

XmlAccessType

-0.44

 utafitiHapana

-0.44



-0.42

 dignidad

-0.38

 resourceCulture

-0.38

UnusedPrivate

-0.37

 متعلقه

-0.37

 sonrisa

-0.36

POSITIVE LOGITS

SharedCtor

0.49

 Normdatei

0.48

guisement

0.43

bootstrapcdn

0.43

varlak

0.42

stellar

0.41

IRQn

0.41

iburg

0.41

biases

0.41

bkz

0.40

Activations Density 0.185%

phrases related to analysis and evaluation of processes or functions

following "phase" or "given"

No Comments

No Known Activations