INDEX

Explanations

references to specific medical conditions, particularly related to immunology and pathology

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

medical suffixes and roots

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

Autoritní

-0.38

arrêter

-0.34

occuper

-0.34

 bolí

-0.33

ocimiento

-0.31

 booking

-0.31

 fédé

-0.31

んですよね

-0.31

 composición

-0.30

attendre

-0.30

POSITIVE LOGITS

RegressionTest

0.71

transQ

0.65

참고

0.64

awaiter

0.60

IANGLES

0.60

WriteTagHelper

0.60

parsedMessage

0.59

aarrggbb

0.58

 consultato

0.57

 InputDecoration

0.57

Activations Density 0.237%

references to specific medical conditions, particularly related to immunology and pathology

medical suffixes and roots

No Comments

No Known Activations