INDEX

Explanations

expressions of opinion or conclusion statements in discussions or analyses

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

featureID

-0.55

 régal

-0.48

нгред

-0.47

 SEDS

-0.45

bootstrapcdn

-0.45

tagext

-0.44

LabelTagHelper

-0.44

敺

-0.44

 tuyo

-0.44

 Regents

-0.43

POSITIVE LOGITS

 justifying

0.41

 nakalista

0.40

 argument

0.39

 justified

0.38

 שוליים

0.38

 gezegd

0.37

arXiv

0.37

 arguments

0.37

 convincing

0.37

XmlAccessorType

0.36

Activations Density 1.313%

expressions of opinion or conclusion statements in discussions or analyses

No Comments

No Known Activations