INDEX

Explanations

terms associated with structures, systems, and frameworks in various contexts

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

levels of, sense of, amounts of, periods between

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

binaan

-0.45

seamnă

-0.44

 guma

-0.43

katapos

-0.41

shutil

-0.41

ientras

-0.40

isError

-0.40

ambién

-0.40

getPoint

-0.39

 Velasco

-0.39

POSITIVE LOGITS

 reminiscent

0.50

Rüyada

0.47

 whose

0.47

 that

0.46

 olyan

0.45

ThroughAttribute

0.44

 StatefulWidget

0.44

ETHING

0.43

cury

0.43

RVA

0.43

Activations Density 0.263%

terms associated with structures, systems, and frameworks in various contexts

levels of, sense of, amounts of, periods between

No Comments

No Known Activations