INDEX

Explanations

restaurants and dining experiences

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

 protoimpl

-0.47

protoimpl

-0.44

ftagPool

-0.37

fæ

-0.36

 learning

-0.36

 phase

-0.36

 material

-0.35

 volatile

-0.35

FunctionFlags

-0.35

dagangan

-0.34

POSITIVE LOGITS

 restaurant

0.85

restaurant

0.79

 restaurants

0.79

 eatery

0.76

 Restaurant

0.76

Restaurant

0.74

 restaurante

0.72

 eateries

0.71

ร้าน

0.68

 Restaurants

0.68

Activations Density 0.027%

restaurants and dining experiences

No Comments

No Known Activations