INDEX

Explanations

intensifiers or modifiers that emphasize quantity or degree

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

Followed by a quantity word

np_acts-logits-general · gemini-2.0-flash

so many

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DebuggerNonUser

-0.82

 متعلقه

-0.73

 onCancelled

-0.71

πως

-0.67

Portale

-0.65

ГЛА

-0.62

 Roskov

-0.61

 ervan

-0.60

참고

-0.59

 AssemblyTitle

-0.59

POSITIVE LOGITS

 fewer

0.74

 some

0.64

Ữ

0.62

 many

0.59

 every

0.57

 several

0.56

 little

0.55

Izvori

0.55

úmer

0.53

opter

0.53

Activations Density 0.343%

intensifiers or modifiers that emphasize quantity or degree

Followed by a quantity word

so many

No Comments

No Known Activations