INDEX

Explanations

conditional phrases and hypothetical statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

jsdelivr

-0.63

sistors

-0.57

NOPQRST

-0.56

GridItem

-0.54

spesies

-0.53

 Somme

-0.53

AlterField

-0.52

nosaurus

-0.51

prefixer

-0.50

mphony

-0.50

POSITIVE LOGITS

 would

1.29

would

1.20

 Would

1.19

Would

1.17

 WOULD

1.16

 wouldn

1.03

 Wouldn

0.99

 wäre

0.90

wouldn

0.90

 gdyby

0.88

Activations Density 0.521%

conditional phrases and hypothetical statements

No Comments

No Known Activations