INDEX

Explanations

references to dining experiences and food consumption

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

kissen

-0.45

conteudo

-0.43

 Task

-0.43

ereza

-0.42

 Integrity

-0.41

 nevoie

-0.40

 verdades

-0.40

mannschaft

-0.39

 evidence

-0.39

 MASTER

-0.39

POSITIVE LOGITS

 restaurants

0.66

 restaurant

0.57

 Restaurants

0.55

restaurants

0.52

 restaurantes

0.51

 abroad

0.50

 ristorante

0.50

Restaurants

0.49

 Restaurant

0.48

dining

0.47

Activations Density 0.017%

references to dining experiences and food consumption

No Comments

No Known Activations