INDEX

Explanations

terms related to acceptance, rejection, and evaluations of outcomes or conditions

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

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

 hush

-0.39

arc

-0.35

arc

-0.33

)++;

-0.33

poz

-0.33

ूट

-0.33

𝟸

-0.32

->_

-0.31

utom

-0.31

 כדי

-0.31

POSITIVE LOGITS

RTHOOK

0.61

addCriterion

0.55

CppMethod

0.54

Välislingid

0.53

!*\

0.53

 utafitiHapana

0.52

 leſs

0.50

הערות

0.48

ValueGenerated

0.48

 jspb

0.48

Activations Density 0.059%

terms related to acceptance, rejection, and evaluations of outcomes or conditions

No Comments

No Known Activations