INDEX

Explanations

proper nouns or names related to individuals or places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 Queries

-0.90

 Quail

-0.90

 Quar

-0.88

qli

-0.82

 QUART

-0.81

 Quer

-0.80

QS

-0.80

 Quilt

-0.79

 quilts

-0.79

Qiu

-0.79

POSITIVE LOGITS

Portail

0.65

Christy

0.64

SpringBoot

0.64

ServletContext

0.63

 Christy

0.61

'{@

0.59

 Hamp

0.56

 gelukkig

0.55

ken

0.55

 protoimpl

0.54

Activations Density 1.546%

proper nouns or names related to individuals or places

No Comments

No Known Activations