INDEX

Explanations

terms related to scientific experiments and methodologies

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

Comes after symbols/numbers

np_acts-logits-general · gemini-2.0-flash

technical modeling and language

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 Prof

-0.40

 stabilis

-0.36

 mekan

-0.36

 Sargent

-0.36

 actionTypes

-0.36

 Lehrer

-0.35

モト

-0.35

 Luxem

-0.35

Fan

-0.35

fer

-0.35

POSITIVE LOGITS

MigrationBuilder

0.63

 nonUne

0.61

DoubleQuotes

0.58

 invokingState

0.56

AndEndTag

0.54

invokeLater

0.53

tonode

0.53

 propOrder

0.49

TRAILING

0.49

'{@

0.48

Activations Density 1.458%

terms related to scientific experiments and methodologies

Comes after symbols/numbers

technical modeling and language

No Comments

No Known Activations