INDEX

Explanations

variations of the verb "be" in different contexts

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

Follows the word "be"

np_acts-logits-general · gemini-2.0-flash

be + predicate

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

 itſelf

-1.71

 myſelf

-1.47

 themſelves

-1.35

 Jefus

-1.35

 pleaſure

-1.34

 ſche

-1.33

 ſeveral

-1.30

 purpoſe

-1.28

 ſtill

-1.28

 whoſe

-1.28

POSITIVE LOGITS

 able

1.12

0.90

an

0.87

the

0.78

be

0.75

 seen

0.75

 held

0.73

0.70

 part

0.68

 taken

0.66

Activations Density 0.413%

variations of the verb "be" in different contexts

Follows the word "be"

be + predicate

No Comments

No Known Activations