INDEX

Explanations

instances of specific formatting or tags likely related to structured data or headings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 myſelf

-1.86

Efq

-1.82

 ―――――

-1.81

 Theſe

-1.78

$_"

-1.76

\\

-1.70

^(@)

-1.69

 Monfieur

-1.69

(\<

-1.68

 Houſe

-1.67

POSITIVE LOGITS

2.02

<bos>

1.56

1.24

1.15

1.10

0.99

↵

0.98

and

0.89

0.87

Activations Density 0.132%

instances of specific formatting or tags likely related to structured data or headings

No Comments

No Known Activations