INDEX

Explanations

structured data representations and numerical expressions

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

c1 and c2

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

WebVitals

-0.51

 étoient

-0.48

rdı

-0.48

หน่อย

-0.47

bootstrapcdn

-0.47

 venganza

-0.46

 viento

-0.45

getYear

-0.45

 tatuaje

-0.45

 gorą

-0.45

POSITIVE LOGITS

0.70

0.67

0.58

CI

0.56

Cl

0.55

CL

0.54

Cx

0.54

 Cara

0.53

 Cluster

0.52

CX

0.52

Activations Density 0.054%

structured data representations and numerical expressions

c1 and c2

No Comments

No Known Activations