INDEX

Explanations

terms related to processes and methods of synchronization and alignment

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

Common linking tokens that introduce definitions, explanations, or technical processes in formal or academic text, often appearing at transitions between concepts.

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

 propOrder

-0.40

verwijspagina

-0.39

 AssemblyCulture

-0.37

 StatefulWidget

-0.36

awtextra

-0.34

hbs

-0.34

 Clan

-0.32

 AspNetCore

-0.31

 المعيارى

-0.30

ceus

-0.30

POSITIVE LOGITS

expandindo

0.56

 process

0.54

Autoritní

0.49

ErrUnexpectedEOF

0.49

uminen

0.48

sphase

0.47

 coisa

0.47

MethodManager

0.46

 restore

0.46

 Reſ

0.46

Activations Density 0.297%

terms related to processes and methods of synchronization and alignment

Common linking tokens that introduce definitions, explanations, or technical processes in formal or academic text, often appearing at transitions between concepts.

No Comments

No Known Activations