INDEX

Explanations

phrases that contain the word "to" and its usage in various contexts

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

request to do something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Portály

-0.37

fact

-0.36

 logro

-0.36

Datuak

-0.32

WHEREAS

-0.31

 históricas

-0.31

وعة

-0.31

 campamento

-0.31

 facility

-0.31

Portale

-0.31

POSITIVE LOGITS

請

0.64

 فريبيس

0.61

AddTagHelper

0.59

.*")]

0.57

Request

0.57

请

0.56

ให

0.56

 Requests

0.55

_("

0.55

 please

0.55

Activations Density 0.009%

phrases that contain the word "to" and its usage in various contexts

request to do something

No Comments

No Known Activations