INDEX

Explanations

terms related to technical or programming concepts, particularly regarding configuration and responses

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

preceding adjectives in other languages

np_acts-logits-general · gemini-2.0-flash

adverbs and adverbial endings

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

httphttps

-0.88

StoryboardSegue

-0.85

 мәкал

-0.83

Personensuche

-0.83

AccessorTable

-0.82

Vidite

-0.82

-0.78

ConfigureAwait

-0.77

 onAnimation

-0.77

 ProtoMessage

-0.77

POSITIVE LOGITS

mente

0.74

 ביותר

0.71

 zaidi

0.67

ๆ

0.65

 للغاية

0.61

 nhất

0.58

MENTE

0.57

amente

0.56

ly

0.54

 جدًا

0.53

Activations Density 0.055%

terms related to technical or programming concepts, particularly regarding configuration and responses

preceding adjectives in other languages

adverbs and adverbial endings

No Comments

No Known Activations