INDEX

Explanations

structured text elements or formatting cues in documents

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

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

 betweenstory

-0.65

migrationBuilder

-0.58

 interactiva

-0.56

 miniaturka

-0.56

 obuv

-0.55

enderror

-0.54

 <=",

-0.54

 extranjera

-0.54

AnchorTagHelper

-0.53

ագրություններ

-0.52

POSITIVE LOGITS

:^(

0.48

 Platon

0.40

 initComponents

0.40

 unifor

0.39

ViewFeatures

0.39

 getConfig

0.36

 flot

0.35

 volu

0.35

 camb

0.35

 volant

0.35

Activations Density 0.000%

structured text elements or formatting cues in documents

No Comments

No Known Activations