INDEX

Explanations

elements related to calculations and parameters within statistical and mathematical context

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

Follows instructions or explanations ("to")

np_acts-logits-general · gemini-2.0-flash

to achieve a goal

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

 autorytatywna

-0.94

Eſ

-0.76

Efq

-0.75

RSpec

-0.73

 Reſ

-0.70

 Shakspeare

-0.69

 ―――――

-0.69

 مشين

-0.68

ſelf

-0.66

 Theſe

-0.66

POSITIVE LOGITS

 must

0.79

서는

0.75

 eerst

0.72

 primero

0.64

must

0.63

 bisogna

0.60

use

0.59

 first

0.59

 надо

0.59

 firstly

0.59

Activations Density 0.300%

elements related to calculations and parameters within statistical and mathematical context

Follows instructions or explanations ("to")

to achieve a goal

No Comments

No Known Activations