INDEX

Explanations

occurrences of double underscores typically used in Python for special methods and attributes

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

verwijspagina

-0.74

ể

-0.67

tv

-0.60

orithmic

-0.58

 propOrder

-0.58

Fak

-0.56

itemize

-0.54

∅

-0.54

stination

-0.54

FormTagHelper

-0.54

POSITIVE LOGITS

__

2.71

(__

1.99

.__

1.91

(__

1.85

 (!__

1.63

)__

1.58

/__

1.57

*__

1.57

[__

1.42

::__

1.40

Activations Density 0.124%

occurrences of double underscores typically used in Python for special methods and attributes

No Comments

No Known Activations