INDEX

Explanations

references to apartments in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 InputDecoration

-0.76

BeginContext

-0.71

 doInBackground

-0.68

jména

-0.64

Hochspringen

-0.63

dij

-0.62

rungsseite

-0.60

ServiceModel

-0.59

āda

-0.59

tichetta

-0.58

POSITIVE LOGITS

 apartment

1.12

 Apartment

0.99

 apartments

0.98

Apartment

0.91

apartment

0.90

Apartments

0.77

 Apartments

0.74

 APART

0.71

 Kome

0.70

 accommodations

0.67

Activations Density 0.066%

references to apartments in various contexts

No Comments

No Known Activations