INDEX

Explanations

terms related to database operations and identifiers

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

AnimationsModule

-0.69

parsedMessage

-0.62

 surla

-0.61

ImageContext

-0.61

withIdentifier

-0.60

﹧

-0.60

 ThemeData

-0.60

 TextAlign

-0.60

 queſta

-0.60

<unused68>

-0.59

POSITIVE LOGITS

 pää

0.38

<eos>

0.35

0.34

 Independencia

0.34

Enllaces

0.33

↵↵

0.32

<h3>

0.32

udd

0.32

<h2>

0.32

ഔ

0.32

Activations Density 0.013%

terms related to database operations and identifiers

No Comments

No Known Activations