INDEX

Explanations

citations and references in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.hook_mlp_out

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

《

-1.27

『

-1.23

《

-1.16

-1.09

-1.07

『

-1.07

‘

-1.05

-1.02

«

-0.99

“

-0.97

POSITIVE LOGITS

 resourceCulture

0.69

 Theſe

0.69

//});

0.67

+#+#

0.67

ówno

0.62

ANTLR

0.61

Aiheesta

0.61

<bos>

0.60

 purpoſe

0.60

IVEREF

0.59

Activations Density 0.118%

citations and references in the text

No Comments

No Known Activations