INDEX

Explanations

elements related to blog interactions and personal reflections

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

First-person present perfect continuous tense

np_acts-logits-general · gemini-2.0-flash

finished up recently

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

彼らは

-0.59

 saiba

-0.59

当时的

-0.56

 ním

-0.56

誰でも

-0.56

 suffit

-0.54

一辈子

-0.54

みてください

-0.54

 automaticamente

-0.54

當時

-0.54

POSITIVE LOGITS

 finally

0.94

finally

0.91

 FINALLY

0.84

 Finally

0.77

 finishing

0.77

Finally

0.76

andaag

0.74

 procrastin

0.73

FINALLY

0.73

 yesterday

0.72

Activations Density 0.470%

elements related to blog interactions and personal reflections

First-person present perfect continuous tense

finished up recently

No Comments

No Known Activations