INDEX

Explanations

elements related to characters and their relationships in a narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Josephus

-0.40

 Bulldog

-0.39

 Bully

-0.39

Tikang

-0.36

FieldNumber

-0.36

 Juifs

-0.35

 jellemző

-0.35

 sensazione

-0.35

 verbess

-0.35

 totaal

-0.35

POSITIVE LOGITS

uxxxx

0.50

 незавершена

0.49

Personendaten

0.48

 infiltrated

0.46

ArrowToggle

0.46

pexpr

0.46

 CreateTagHelper

0.45

httphttps

0.45

 conspira

0.44

 infiltration

0.42

Activations Density 0.641%

elements related to characters and their relationships in a narrative

No Comments

No Known Activations