INDEX

Explanations

references to television shows and episodes

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 NSCoder

-0.51

articolo

-0.47

rungsseite

-0.47

 Wiktionnaire

-0.47

 Eindruck

-0.46

 AssemblyProduct

-0.42

 sarili

-0.42

bigli

-0.42

 homeschool

-0.40

 ainfi

-0.40

POSITIVE LOGITS

WebElementEntity

0.39

posedge

0.38

 ویکی‌پدیای

0.38

exp

0.36

Stream

0.36

WithIOException

0.36

ransition

0.34

허

0.34

lock

0.34

 explan

0.34

Activations Density 0.143%

references to television shows and episodes

No Comments

No Known Activations