INDEX

Explanations

references to time-related concepts such as months, years, and days

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

month and year manipulation

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

 للاسماء

-0.65

principalTable

-0.65

DockStyle

-0.57

PerformLayout

-0.56

GraphicsUnit

-0.55

elemField

-0.55

ViewFeatures

-0.51

dain

-0.50

saraba

-0.49

UpInside

-0.49

POSITIVE LOGITS

 broder

0.35

glLoadIdentity

0.34

owiec

0.33

quele

0.33

æder

0.32

 vêtement

0.32

Aluno

0.31

XmlAccessorType

0.31

Merkmale

0.31

OMEM

0.31

Activations Density 0.012%

references to time-related concepts such as months, years, and days

month and year manipulation

No Comments

No Known Activations