INDEX

Explanations

items indicating a reduction or removal of something

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

Section or document headers that introduce new content blocks, typically appearing at document beginnings or after metadata sections.

eleuther_acts_top20 · claude-3-7-sonnet-20250219 Triggered by @lu-christina

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

UnusedPrivate

-0.58

MLLoader

-0.57

出版年

-0.55

sidemargin

-0.54

ReusableCell

-0.53

 propOrder

-0.53

StoryboardSegue

-0.52

oneofs

-0.48

withIdentifier

-0.47

JspWriter

-0.47

POSITIVE LOGITS

<eos>

0.54

 EnglishChoose

0.38

Clean

0.38

 clean

0.36

 Karo

0.35

 Clean

0.35

</blockquote>

0.35

FIM

0.35

 borer

0.35

Boring

0.35

Activations Density 0.031%

items indicating a reduction or removal of something

Section or document headers that introduce new content blocks, typically appearing at document beginnings or after metadata sections.

No Comments

No Known Activations