INDEX

Explanations

phrases indicating relationships and connections between concepts or subjects

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

Relationships and actions

np_max-act · gemini-2.0-flash Triggered by @johnny

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_1m/average_l0_107

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

1,048,576

Data Type

float32

Hook Name

blocks.12.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

 dijual

-0.50

 kehadiran

-0.46

 dragen

-0.44

 kérdés

-0.44

occas

-0.43

 dikeluarkan

-0.43

 wereld

-0.43

 enseignants

-0.42

 trägt

-0.42

 yapılan

-0.41

POSITIVE LOGITS

awtextra

0.92

IntoConstraints

0.85

AddTagHelper

0.81

</thead>

0.77

цезда

0.75

 resourceCulture

0.74

StoreMessageInfo

0.73

AutoresizingMask

0.73

 myſelf

0.72

 transfieras

0.70

Activations Density 0.022%

phrases indicating relationships and connections between concepts or subjects

Relationships and actions

No Comments

No Known Activations