INDEX

Explanations

programming-related terms and phrases that indicate problem-solving actions

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

problem solving or possibility

np_acts-logits-general · gemini-2.0-flash

potential solutions or outcomes

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

Хьажоргаш

-0.58

RegressionTest

-0.55

Tracce

-0.52

ніципалі

-0.51

 Partagez

-0.50

arakhand

-0.49

Taj

-0.48

kazy

-0.47

 Вікі

-0.47

 esempi

-0.47

POSITIVE LOGITS

 solves

1.19

 alleviate

1.15

 solve

1.14

 solved

0.99

 solving

0.95

solve

0.94

解決

0.91

 cure

0.90

 amelior

0.88

 alleviation

0.86

Activations Density 0.391%

programming-related terms and phrases that indicate problem-solving actions

problem solving or possibility

potential solutions or outcomes

No Comments

No Known Activations