INDEX

Explanations

numeric values related to health measurements and statistics

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

EClass

-0.60

flashdata

-0.59

χρι

-0.56

+#+#

-0.55

faßt

-0.54

hí

-0.53

endphp

-0.53

GraphicsUnit

-0.53

panou

-0.52

lampa

-0.52

POSITIVE LOGITS

 correlation

2.22

 correlations

1.98

 Correlation

1.89

Correlation

1.80

correlation

1.80

 Correlations

1.70

 correl

1.53

 correlate

1.50

 correlated

1.49

 relationship

1.32

Activations Density 0.957%

numeric values related to health measurements and statistics

No Comments

No Known Activations