INDEX

Explanations

terms related to technical specifications and programming concepts

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

MODE context process

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

 surla

-0.72

rungsseite

-0.68

stateProvider

-0.63

IBarButtonItem

-0.61

期刊论文

-0.60

 Signalez

-0.59

évaluateur

-0.59

agisse

-0.59

transQ

-0.58

$_"

-0.58

POSITIVE LOGITS

MODE

0.53

WritableDatabase

0.51

Mode

0.40

StandardCharsets

0.40

 Innenstadt

0.35

 dictadura

0.34

mode

0.34

 meilleurs

0.34

 Konstruktion

0.34

MigrationBuilder

0.34

Activations Density 0.000%

terms related to technical specifications and programming concepts

MODE context process

No Comments

No Known Activations