INDEX

Explanations

marked sections of text, such as dividers or introductory phrases (indicated by <bos>), along with certain patterns or terms related to data analysis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

MessageTagHelper

-0.72

MongoClient

-0.54

zzleHttp

-0.54

PMailer

-0.47

 bientôt

-0.47

nehm

-0.46

 AssemblyCompany

-0.46

Captcha

-0.45

Zitat

-0.45

PORARY

-0.45

POSITIVE LOGITS

 AssemblyCulture

0.92

 usual

0.72

Sucesor

0.72

üb

0.71

FunctionFlags

0.67

Normally

0.66

 normalerweise

0.65

tagHelperRunner

0.65

 autorytatywna

0.65

Usually

0.64

Activations Density 0.734%

marked sections of text, such as dividers or introductory phrases (indicated by <bos>), along with certain patterns or terms related to data analysis

No Comments

No Known Activations