INDEX

Explanations

references to the verb "be" in various contexts

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

be

np_max-act-logits · gemini-2.0-flash Triggered by @johnny

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DoubleQuotes

-0.49

enough

-0.48

sufficient

-0.48

 enough

-0.48

 want

-0.47

[]

-0.47

 Calculate

-0.45

าล

-0.45

ADELPHIA

-0.44

 attempt

-0.43

POSITIVE LOGITS

:✨

0.95

 otomatig

0.85

ReusableCell

0.85

 مرئيه

0.84

 lenker

0.81

adaptiveStyles

0.80

 LUMP

0.78

twimg

0.77

CloseOperation

0.77

rungsseite

0.77

Activations Density 0.128%

references to the verb "be" in various contexts

be

No Comments

No Known Activations