INDEX

Explanations

references to language proficiency and translation contexts

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

$_"

-0.67

%";

-0.65

%");

-0.61

Искәрмәләр

-0.60

izability

-0.60

сылкі

-0.59

__":

-0.58

 proportionality

-0.58

."));

-0.57

 Alamy

-0.57

POSITIVE LOGITS

 languages

0.67

 pinulongan

0.64

 AssemblyCulture

0.58

 language

0.57

/******/

0.55

 English

0.55

StandardCharsets

0.53

InjectAttribute

0.52

 Meksiku

0.52

 langue

0.51

Activations Density 0.125%

references to language proficiency and translation contexts

No Comments

No Known Activations