INDEX

Explanations

phrases expressing difficulty or complexity

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

"difficult to" followed by a verb

np_acts-logits-general · gemini-2.0-flash

difficult to do

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AndEndTag

-0.80

abestanden

-0.73

jsonwebtoken

-0.72

MessageTagHelper

-0.70

 Италијани

-0.69

ModelAdmin

-0.69

AccessorTable

-0.69

webElementXpaths

-0.68

UpInside

-0.67

 springfox

-0.66

POSITIVE LOGITS

 task

0.85

 tâche

0.81

 difficult

0.80

 Difficult

0.79

Difficult

0.78

 tarefa

0.74

difficult

0.71

 Schwier

0.71

 impossible

0.69

 tough

0.69

Activations Density 0.088%

phrases expressing difficulty or complexity

"difficult to" followed by a verb

difficult to do

No Comments

No Known Activations