INDEX

Explanations

actions and verbs indicating observation or interaction

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

and

-0.44

rimos

-0.44

性和

-0.43

CppMethod

-0.43

この日

-0.43

 समीक्षक

-0.41

差し

-0.40

 Station

-0.40

 Ostat

-0.40

 ostat

-0.38

POSITIVE LOGITS

IntoConstraints

0.63

KommentareTeilen

0.63

 invokingState

0.51

aarrggbb

0.50

脚注の使い方

0.49

+#+#

0.48

 חיצוניים

0.48

ConstraintMaker

0.46

 BoxFit

0.46

ImageContext

0.44

Activations Density 0.059%

actions and verbs indicating observation or interaction

No Comments

No Known Activations