INDEX

Explanations

mathematical symbols and notation, particularly those used in equations and formulas

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

\llbracket \langle \rangle \rrbracket

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

 Schloß

-0.42

 Westen

-0.42

 Schöne

-0.35

 Noviembre

-0.34

 whenever

-0.34

by

-0.34

 Neubau

-0.34

 Bersama

-0.34

 charité

-0.34

 called

-0.33

POSITIVE LOGITS

parsedMessage

0.96

RenderAtEndOf

0.84

 صوتيه

0.74

 ſondern

0.73

ésultats

0.73

 ſei

0.73

yntaxException

0.73

⪜

0.72

VYMaps

0.72

 ddelweddau

0.71

Activations Density 0.080%

mathematical symbols and notation, particularly those used in equations and formulas

\llbracket \langle \rangle \rrbracket

No Comments

No Known Activations