INDEX

Explanations

references to the concept of "ultimate" or "ultimate goals."

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

ckså

-0.62

actéristi

-0.56

rollers

-0.55

 للمعارف

-0.54

 Snakes

-0.52

 Craw

-0.51

étoit

-0.51

inery

-0.51

 Könige

-0.50

kiyor

-0.50

POSITIVE LOGITS

 Ultimate

0.92

Ultimate

0.88

 ultimate

0.85

ultimate

0.84

ULT

0.76

 ulti

0.67

 Ultima

0.64

final

0.62

 FINAL

0.62

 Altman

0.58

Activations Density 0.006%

references to the concept of "ultimate" or "ultimate goals."

No Comments

No Known Activations