INDEX

Explanations

words related to exceptions and exclusivity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 Marlene

-0.69

PMailer

-0.67

WebVitals

-0.66

 Barnabas

-0.65

windig

-0.64

*/;

-0.64

viewDidLoad

-0.63

IsMutable

-0.62

Personendaten

-0.62

Gön

-0.61

POSITIVE LOGITS

EX

2.18

EX

2.13

Ex

2.10

Ex

2.03

ex

2.00

ex

1.83

dex

1.71

lex

1.69

rex

1.66

pex

1.65

Activations Density 2.534%

words related to exceptions and exclusivity

No Comments

No Known Activations