INDEX

Explanations

references to scientific data and methodologies

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

scientific terms and procedures

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

 detailsLots

-0.40

 justamente

-0.37

 justement

-0.36

 ferons

-0.36

stdafx

-0.36

 lembra

-0.35

argb

-0.34

Remember

-0.34

jä

-0.34

 Segel

-0.33

POSITIVE LOGITS

AddTagHelper

0.68

MigrationBuilder

0.66

AndEndTag

0.56

CloseOperation

0.53

 AssemblyTitle

0.53

 CreateTagHelper

0.52

]")]

0.52

ConstraintMaker

0.51

VersionUID

0.51

esModule

0.51

Activations Density 0.094%

references to scientific data and methodologies

scientific terms and procedures

No Comments

No Known Activations