INDEX

Explanations

references to quantities and comparisons that denote relationships or conditions in data

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

Izvori

-0.46

RuleContext

-0.39

bait

-0.39

Kalb

-0.38

IDS

-0.38

Bead

-0.38

vyt

-0.37

drill

-0.37

 Vandal

-0.36

riors

-0.35

POSITIVE LOGITS

 respectively

0.91

 respectivement

0.76

respectively

0.73

 respective

0.72

 respectivamente

0.71

pective

0.68

yntaxException

0.63

WindowConstants

0.59

 noqa

0.56

GOTREF

0.56

Activations Density 0.214%

references to quantities and comparisons that denote relationships or conditions in data

No Comments

No Known Activations