INDEX

Explanations

punctuation and indicators of speech or thought

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

Follows punctuation marks and sentence breaks

np_acts-logits-general · gemini-2.0-flash

appears logical or evident

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

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

 NSCoder

-0.92

AndEndTag

-0.81

Personensuche

-0.79

GEBURTSDATUM

-0.76

ंदीखरीदारी

-0.70

جغرافيا

-0.70

AsStream

-0.70

ReusableCell

-0.69

WriteBarrier

-0.68

 للمعارف

-0.67

POSITIVE LOGITS

 Apparently

1.23

Apparently

1.21

 apparently

1.14

なるほど

1.11

 Sounds

1.10

Sounds

1.10

 Clearly

1.09

 obviously

1.09

 clearly

1.08

 Evidently

1.08

Activations Density 0.376%

punctuation and indicators of speech or thought

Follows punctuation marks and sentence breaks

appears logical or evident

No Comments

No Known Activations