INDEX

Explanations

key legal terms or phrases related to official documents or communications

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

user input and context

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

MathML

-0.50

 Dessa

-0.44

 vzor

-0.44

颊

-0.43

เปล

-0.41

gründ

-0.41

attuale

-0.41

 nearly

-0.40

artigen

-0.40

 Heller

-0.39

POSITIVE LOGITS

CloseOperation

0.58

LookAnd

0.54

ModelSerializer

0.53

Autoritní

0.52

DockStyle

0.51

UseVisualStyle

0.48

TagMode

0.48

참고

0.46

omitempty

0.45

Geplaatst

0.44

Activations Density 0.000%

key legal terms or phrases related to official documents or communications

user input and context

No Comments

No Known Activations