INDEX

Explanations

quantities or counts related to objects or subjects

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

Numbers followed by a noun

np_acts-logits-general · gemini-2.0-flash

numerical counts of items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

enumi

-0.83

 ddelweddau

-0.81

usermodel

-0.66

 istrinya

-0.66

 <=",

-0.65

CodeDom

-0.64

enderror

-0.64

 CreateTagHelper

-0.64

 MainAxisSize

-0.62

 EconPapers

-0.62

POSITIVE LOGITS

 different

1.05

teen

0.76

zehn

0.71

 distinct

0.70

 separate

0.70

 more

0.69

different

0.69

(!)

0.67

TEEN

0.65

不同的

0.63

Activations Density 0.572%

quantities or counts related to objects or subjects

Numbers followed by a noun

numerical counts of items

No Comments

No Known Activations