INDEX

Explanations

phrases related to actions and transitions

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

Followed by "to" or "for"

np_acts-logits-general · gemini-2.0-flash

continuation or progression

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

RenderAtEndOf

-0.80

setRotation

-0.59

TagHelper

-0.59

TERY

-0.58

ReusableCell

-0.57

")[

-0.57

 both

-0.57

balleur

-0.56

ָׁ

-0.56

 regards

-0.55

POSITIVE LOGITS

 виправивши

0.78

انيف

0.66

 يتيمه

0.61

utnant

0.60

numerusform

0.59

AnimationsModule

0.58

rungsseite

0.57

 alphabetically

0.54

0.53

InSection

0.53

Activations Density 0.062%

phrases related to actions and transitions

Followed by "to" or "for"

continuation or progression

No Comments

No Known Activations