INDEX

Explanations

SQL query-related terminology

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

SQL table and database references

np_acts-logits-general · gemini-2.0-flash

SQL query syntax

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

 следова

-0.54

 Consequently

-0.52

Consequently

-0.50

thus

-0.48

 consequently

-0.48

pso

-0.45

Therefore

-0.45

 ever

-0.45

 hence

-0.44

Nu

-0.44

POSITIVE LOGITS

 المعيارى

0.68

ècie

0.67

 فريبيس

0.65

 متعلقه

0.63

 where

0.63

 chrétien

0.62

 smtplib

0.60

horabuena

0.60

ніципалі

0.60

 atheism

0.59

Activations Density 0.125%

SQL query-related terminology

SQL table and database references

SQL query syntax

No Comments

No Known Activations