INDEX

Explanations

textual elements indicating functionality or processes

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

medical research and findings

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

 utafitiHapana

-0.79

 للمعارف

-0.75

المناصب

-0.73

LayoutConstraint

-0.62

 saites

-0.60

 MainAxisSize

-0.58

rítica

-0.56

migrationBuilder

-0.54

 lenker

-0.53

parametrize

-0.51

POSITIVE LOGITS

Val

0.38

 fır

0.35

ById

0.35

ød

0.34

Ỏ

0.33

 Anti

0.32

Boe

0.32

off

0.32

Pfe

0.31

prozess

0.31

Activations Density 0.178%

textual elements indicating functionality or processes

medical research and findings

No Comments

No Known Activations