INDEX

Explanations

ellipses and punctuation marks indicating pauses or trailing thoughts

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

Follows ellipses or incomplete sentences

np_acts-logits-general · gemini-2.0-flash

trailing punctuation

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

NUMX

-0.98

expandindo

-0.96

NewUrlParser

-0.93

 للمعارف

-0.81

 NSCoder

-0.81

 autorytatywna

-0.81

Tikang

-0.80

bootstrapcdn

-0.77

CloseOperation

-0.77

SBATCH

-0.75

POSITIVE LOGITS

↵↵

1.30

↵

0.99

↵↵↵

0.94

<eos>

0.93

The

0.79

↵↵↵↵

0.79

if

0.76

but

0.76

and

0.75

it

0.73

Activations Density 0.113%

ellipses and punctuation marks indicating pauses or trailing thoughts

Follows ellipses or incomplete sentences

trailing punctuation

No Comments

No Known Activations