INDEX

Explanations

terms related to performance metrics and comparisons between groups

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

Preceding words followed by prepositions

np_acts-logits-general · gemini-2.0-flash

knowledge of

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

AndEndTag

-0.54

hos

-0.45

 globo

-0.44

bewerken

-0.43

ti

-0.42

kem

-0.41

smo

-0.40

rez

-0.40

Yb

-0.40

sem

-0.40

POSITIVE LOGITS

 בתח

1.15

 inom

0.97

ด้าน

0.91

 concernant

0.85

 בנושא

0.85

 về

0.84

tvguidetime

0.81

 riguardo

0.81

 domaines

0.79

in

0.78

Activations Density 0.521%

terms related to performance metrics and comparisons between groups

Preceding words followed by prepositions

knowledge of

No Comments

No Known Activations