INDEX

Explanations

numerical representations or metrics

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

Numbers or values inside parentheses

np_acts-logits-general · gemini-2.0-flash

jälkeen compared joten epä pitä

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

tvguidetime

-0.62

$_(

-0.62

ロウィン

-0.61

 好文分享

-0.60

лтамалар

-0.60

 intptr

-0.59

ViewFeatures

-0.59

<unused41>

-0.59

[@BOS@]

-0.59

<unused21>

-0.59

POSITIVE LOGITS

 epä

0.42

 päät

0.40

 pitä

0.39

 jäm

0.39

 despite

0.37

 compared

0.37

 kaksi

0.37

Despite

0.37

 jälkeen

0.36

 joten

0.35

Activations Density 1.084%

numerical representations or metrics

Numbers or values inside parentheses

jälkeen compared joten epä pitä

No Comments

No Known Activations