INDEX

Explanations

phrases describing methods or approaches to doing something

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

+#+#

-0.85

findpost

-0.74

BeginContext

-0.74

انيف

-0.64

AsUp

-0.62

migrationBuilder

-0.62

 nargin

-0.59

databind

-0.59

Rujukan

-0.58

RegistryLite

-0.57

POSITIVE LOGITS

 manner

0.77

way

0.62

 Weise

0.61

 modo

0.60

 disponibilités

0.59

的方式

0.58

 extent

0.56

posób

0.56

 pewno

0.55

 način

0.55

Activations Density 0.470%

phrases describing methods or approaches to doing something

No Comments

No Known Activations