INDEX

Explanations

references to binary or binomial concepts

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

The token "bin" or "Bin"

np_acts-logits-general · gemini-2.0-flash

binning, binomial, binned, binaries, bin Laden

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

 jScrollPane

-0.63

endregion

-0.55

Kanpo

-0.55

 Leicht

-0.52

iança

-0.51

 chi̍t

-0.48

UserScript

-0.47

NameValuePair

-0.47

 capo

-0.46

هرست

-0.46

POSITIVE LOGITS

bin

0.97

Bin

0.93

 bins

0.84

BIN

0.83

numerusform

0.77

Bin

0.74

 bags

0.74

bin

0.73

bag

0.73

BIN

0.73

Activations Density 0.087%

references to binary or binomial concepts

The token "bin" or "Bin"

binning, binomial, binned, binaries, bin Laden

No Comments

No Known Activations