INDEX

Explanations

sentences that convey strong emotions or impactful statements

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 propOrder

-0.61

 surla

-0.58

Décès

-0.57

parsedMessage

-0.57

LikeLike

-0.55

 nonUne

-0.55

 Wicidata

-0.54

MLLoader

-0.52

الحياه

-0.52

 EconPapers

-0.52

POSITIVE LOGITS

 truly

0.29

rinha

0.28

unico

0.28

none

0.28

intervento

0.28

 bacio

0.28

truly

0.27

ļa

0.27

ági

0.27

Activations Density 0.015%

sentences that convey strong emotions or impactful statements

No Comments

No Known Activations