INDEX

Explanations

words and phrases indicating relationships, particularly in contexts involving actions and connections

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

ური

-0.63

 veta

-0.61

iade

-0.58

 وتسجيلات

-0.54

MSR

-0.53

zoic

-0.53

 giro

-0.52

งาม

-0.51

liothèque

-0.51

közi

-0.51

POSITIVE LOGITS

 WaitForSeconds

0.74

windowFixed

0.70

+#+#

0.69

StoryboardSegue

0.65

Hochspringen

0.62

szön

0.62

 EdgeInsets

0.60

TestingModule

0.58

LookAnd

0.58

 kepad

0.57

Activations Density 0.627%

words and phrases indicating relationships, particularly in contexts involving actions and connections

No Comments

No Known Activations