INDEX

Explanations

structured elements within written documents

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

paragraph and sentences

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

Negative Logits

+#+

-0.55

OrNil

-0.48

RTEE

-0.46

'{@

-0.41

 TERMIN

-0.41

 hạn

-0.41

얻

-0.40

ukone

-0.40

FunctionFlags

-0.39

uertos

-0.39

POSITIVE LOGITS

 Réponses

0.57

setVerticalGroup

0.54

 autorytatywna

0.52

AutoScale

0.51

 CURIAM

0.50

 paragraph

0.49

AccessorTable

0.48

MergeFrom

0.47

RenderAtEndOf

0.45

 redacción

0.44

Activations Density 0.039%

structured elements within written documents

paragraph and sentences

No Known Activations