INDEX

Explanations

expressions related to intensity and qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

 تضيفلها

-0.62

not

-0.52

")[

-0.51

FLICT

-0.51

dula

-0.50

setUse

-0.49

})));

-0.49

atore

-0.49

"))

-0.48

ile

-0.48

POSITIVE LOGITS

 pleaſure

0.80

 greateſt

0.75

 myſelf

0.75

 itſelf

0.74

 Majefty

0.69

 themſelves

0.68

 himſelf

0.68

 againſt

0.67

 purpoſe

0.65

دانشنامهٔ

0.65

Activations Density 0.373%

expressions related to intensity and qualities

No Comments

No Known Activations