INDEX

Explanations

important terms and keywords related to applications, documents, and formal processes

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

dates, rules, computer terms

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

 Simons

-0.32

 pedras

-0.30

 respective

-0.30

 variety

-0.30

LError

-0.29

Po

-0.29

 faixa

-0.28

 risks

-0.28

 references

-0.28

クロ

-0.27

POSITIVE LOGITS

astéroïdes

0.74

setVerticalGroup

0.72

hisattva

0.70

IndentedString

0.70

IContainer

0.69

ougars

0.69

AnchorTagHelper

0.69

windowFixed

0.67

 CURIAM

0.67

évaluateur

0.63

Activations Density 0.002%

important terms and keywords related to applications, documents, and formal processes

dates, rules, computer terms

No Comments

No Known Activations