INDEX

Explanations

instances of complexity or complications 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 @ 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

 си

-0.59

✭✭

-0.55

 intent

-0.54

dė

-0.52

 adult

-0.51

 ве

-0.50

歇

-0.50

diac

-0.50

 sustained

-0.50

 ле

-0.50

POSITIVE LOGITS

 tricky

1.53

 pricey

1.42

 complicado

1.20

 dodgy

1.18

 trendy

1.16

complicated

1.15

 shady

1.13

 complicada

1.13

trendy

1.12

 lengthy

1.11

Activations Density 0.113%

instances of complexity or complications in various contexts

No Comments

No Known Activations