INDEX

Explanations

mathematical notation and terminology related to sequences and indexing

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

Superscripted index (i, th, st, nd, rd)

np_acts-logits-general · gemini-2.0-flash

nth terms and positional indices

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

ten

-0.56

for

-0.55

ANDUM

-0.53

 five

-0.50

an

-0.48

 about

-0.47

 fifteen

-0.46

 which

-0.46

-0.45

POSITIVE LOGITS

th

1.02

AddTagHelper

0.90

'{@

0.87

principalColumn

0.84

 iſt

0.84

AndEndTag

0.84

 firſt

0.82

nth

0.81

 greateſt

0.81

 تضيفلها

0.79

Activations Density 0.164%

mathematical notation and terminology related to sequences and indexing

Superscripted index (i, th, st, nd, rd)

nth terms and positional indices

No Comments

No Known Activations