INDEX

Explanations

references to databases and related terminology

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

database or db contexts

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

parsedMessage

-0.58

artifactId

-0.58

ansha

-0.55

isit

-0.54

 strerror

-0.54

eid

-0.54

 myſelf

-0.54

heil

-0.53

permitAll

-0.53

WEBPACK

-0.52

POSITIVE LOGITS

db

1.19

db

1.19

DB

1.12

DB

1.07

Db

0.90

Db

0.90

FromDB

0.82

dbs

0.79

dbs

0.77

 database

0.76

Activations Density 0.012%

references to databases and related terminology

database or db contexts

No Comments

No Known Activations