INDEX

Explanations

references to error codes and programming constructs relevant to system operations and function calls

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

RegressionTest

-0.81

 Roskov

-0.64

 dAtA

-0.60

хьтан

-0.59

انيف

-0.58

 يتيمه

-0.58

脚注の使い方

-0.58

 Wikimedijinoj

-0.57

Rhestr

-0.57

 nakalista

-0.57

POSITIVE LOGITS

ailer

0.47

icolor

0.46

akse

0.46

itul

0.46

eradish

0.46

$_['

0.45

 CreateTagHelper

0.45

adone

0.43

werf

0.43

,...,

0.43

Activations Density 0.322%

references to error codes and programming constructs relevant to system operations and function calls

No Comments

No Known Activations