INDEX

Explanations

connective words and phrases that indicate relationships or inclusions in arguments or theories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 onAnimation

-0.62

لیس

-0.61

 polio

-0.61

 calcite

-0.56

Efq

-0.56

 Tripoli

-0.56

 ſche

-0.56

LEGGI

-0.56

 Thra

-0.56

 iſt

-0.55

POSITIVE LOGITS

℗

0.66

 *)__

0.62

CodeAttribute

0.61

 насељу

0.59

 Wicidata

0.57

 Normdatei

0.57

Diweddarwch

0.57

 ואת

0.56

AlterField

0.56

Jîn

0.54

Activations Density 0.236%

connective words and phrases that indicate relationships or inclusions in arguments or theories

No Comments

No Known Activations