INDEX

Explanations

abstract concepts and relationships, particularly in mathematical or theoretical contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Wicidata

-0.86

 defaultstate

-0.61

cshtml

-0.61

writeFieldEnd

-0.60

[])

-0.59

RegressionTest

-0.56

*/

-0.56

AxisAlignment

-0.56

stander

-0.55

épend

-0.54

POSITIVE LOGITS

 characteristics

0.97

characteristics

0.81

 relevance

0.72

 features

0.70

 tendency

0.69

 implications

0.68

 Characteristics

0.68

 significance

0.68

 appeal

0.68

 resemblance

0.67

Activations Density 0.855%

abstract concepts and relationships, particularly in mathematical or theoretical contexts

No Comments

No Known Activations