INDEX

Explanations

references to trains and their related terminology

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

PPA

-0.57

../../../

-0.56

ximity

-0.54

 fundos

-0.54

 nakalista

-0.54

ValueStyle

-0.54

ronpa

-0.53

MemoryWarning

-0.52

 Chwiliwch

-0.51

 Margo

-0.51

POSITIVE LOGITS

 Train

1.04

 train

0.96

Train

0.95

 TRAIN

0.95

train

0.87

 Trains

0.77

 trains

0.76

TRAIN

0.73

Trains

0.66

 istore

0.65

Activations Density 0.005%

references to trains and their related terminology

No Comments

No Known Activations