INDEX

Explanations

detailed analyses and accounts of various subjects, often emphasizing thorough evaluations or explorations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

Behavior

-0.38

SharedCtor

-0.38

moveToFirst

-0.38

Secret

-0.35

 hero

-0.35

Topic

-0.34

 advisors

-0.34

PYX

-0.34

MathML

-0.33

pez

-0.33

POSITIVE LOGITS

 depiction

0.59

 description

0.58

 BoxFit

0.56

 descrição

0.53

 описание

0.52

 portrayal

0.51

 exploration

0.50

 asupra

0.50

 descripción

0.50

+#+#

0.49

Activations Density 1.180%

detailed analyses and accounts of various subjects, often emphasizing thorough evaluations or explorations

No Comments

No Known Activations