INDEX

Explanations

references to size or magnitude in various contexts

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

"large" or "big"

np_acts-logits-general · gemini-2.0-flash

big or large magnitude

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

ReusableCell

-0.58

SuppressLint

-0.53

 convertView

-0.50

Mate

-0.50

PRWEB

-0.47

 conditions

-0.47

pielen

-0.46

 крайней

-0.46

τως

-0.44

>{@

-0.44

POSITIVE LOGITS

 sized

1.01

 amounts

1.00

 quantities

0.88

amounts

0.87

Huge

0.86

 chunks

0.86

sized

0.82

 للاسماء

0.82

 size

0.81

 scale

0.79

Activations Density 0.315%

references to size or magnitude in various contexts

"large" or "big"

big or large magnitude

No Comments

No Known Activations