INDEX

Explanations

terms related to organizations, studies, or evaluations in various contexts

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

company, study, disease, party

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

󠁢

-0.36

 Amérique

-0.35

терак

-0.35

 nhau

-0.35

Portail

-0.35

Portale

-0.35

 surla

-0.34

dataclass

-0.34

agriculture

-0.34

 inalámbrico

-0.34

POSITIVE LOGITS

 nahilalakip

0.56

ABUL

0.52

 createSlice

0.48

RenderAtEndOf

0.47

Slf

0.46

MigrationBuilder

0.46

Diweddarwch

0.46

mbg

0.45

bular

0.45

ibrator

0.45

Activations Density 0.084%

terms related to organizations, studies, or evaluations in various contexts

company, study, disease, party

No Comments

No Known Activations