INDEX

Explanations

words and phrases that indicate commonality or frequency in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

StoryboardSegue

-0.90

Personendaten

-0.80

makeConstraints

-0.77

+#+#

-0.74

 GenerationType

-0.72

ViewInit

-0.72

AddTagHelper

-0.71

 createSlice

-0.66

出版年

-0.65

 geweest

-0.63

POSITIVE LOGITS

 popular

0.83

 commonly

0.82

 common

0.81

 frequently

0.75

常见

0.68

common

0.67

常见的

0.67

 often

0.67

Popular

0.66

 commonest

0.66

Activations Density 0.491%

words and phrases that indicate commonality or frequency in various contexts

No Comments

No Known Activations