INDEX

Explanations

references to the name "Val."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

 Moro

-0.54

vor

-0.54

DLE

-0.53

Ừ

-0.52

Phyllis

-0.52

ukt

-0.52

 Isis

-0.50

 frig

-0.50

透

-0.49

Ums

-0.49

POSITIVE LOGITS

Val

3.35

Val

3.01

VAL

2.48

val

1.89

VAL

1.61

 Vali

1.51

 Valdés

1.45

 Valer

1.45

 Valdez

1.40

 valet

1.31

Activations Density 0.080%

references to the name "Val."

No Comments

No Known Activations