INDEX

Explanations

phrases related to instructions or guidance

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

"How" followed by "to"

np_acts-logits-general · gemini-2.0-flash

how to instructions

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

astify

-0.88

aarrggbb

-0.87

ंदीखरीदारी

-0.82

 InputDecoration

-0.82

 дописавши

-0.82

 <=",

-0.80

WebVitals

-0.79

StoryboardSegue

-0.78

writeFieldEnd

-0.77

 Seeder

-0.77

POSITIVE LOGITS

to

0.91

can

0.53

 να

0.53

best

0.51

 best

0.49

how

0.47

can

0.47

ecan

0.45

 dirig

0.44

to

0.43

Activations Density 0.062%

phrases related to instructions or guidance

"How" followed by "to"

how to instructions

No Comments

No Known Activations