INDEX

Explanations

references to government actions and their implications

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

Fram

-0.36

 seren

-0.35

 skep

-0.35

HUM

-0.33

 doub

-0.32

 Nere

-0.32

 EXPL

-0.32

 keli

-0.31

Ske

-0.31

 kost

-0.30

POSITIVE LOGITS

RotationOrder

0.74

Personendaten

0.62

BASEPATH

0.62

 OMITBAD

0.59

 actionMode

0.59

sizeCache

0.59

CDCl

0.59

__":

0.58

MigrationBuilder

0.58

таратура

0.56

Activations Density 0.128%

references to government actions and their implications

No Comments

No Known Activations