INDEX

Explanations

comparative phrases involving levels of importance or effectiveness

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

", if" or ", if not"

np_acts-logits-general · gemini-2.0-flash

if not all

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

 CreateTagHelper

-0.77

Enllaços

-0.71

bolista

-0.66

">:

-0.65

StoryboardSegue

-0.65

(;;)

-0.65

-0.64

 autorytatywna

-0.63

__).

-0.61

latego

-0.60

POSITIVE LOGITS

 voire

1.30

if

1.23

if

1.04

甚至是

0.91

甚至

0.90

乃至

0.90

or

0.87

 если

0.86

 hatta

0.86

 bahkan

0.82

Activations Density 0.260%

comparative phrases involving levels of importance or effectiveness

", if" or ", if not"

if not all

No Comments

No Known Activations