INDEX

Explanations

terms related to system design and functionality

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

preceding lists or categories

np_acts-logits-general · gemini-2.0-flash

domains and categories

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

seamnă

-0.52

varlak

-0.47

isSuccess

-0.44

binaan

-0.43

 saja

-0.42

madamente

-0.42

ARROLL

-0.42

Autocomplete

-0.41

ſelben

-0.40

 Италијани

-0.40

POSITIVE LOGITS

 reminiscent

0.50

 olyan

0.49

 similar

0.48

providedIn

0.46

 صوتيه

0.45

 هایی

0.44

 BorderSide

0.43

 that

0.43

 such

0.43

 consultato

0.42

Activations Density 0.956%

terms related to system design and functionality

preceding lists or categories

domains and categories

No Comments

No Known Activations