INDEX

Explanations

instances and references related to ethical principles and frameworks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

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

transQ

-0.71

 BoxFit

-0.67

yntaxException

-0.66

 kasarigan

-0.59

 queſta

-0.58

antMatchers

-0.57

 CascadeType

-0.57

#+#

-0.57

pym

-0.56

يكب

-0.56

POSITIVE LOGITS

findpost

0.38

wußt

0.36

 Zentral

0.34

Every

0.33

 <>",

0.32

Define

0.31

 päät

0.30

chowy

0.29

 fédéral

0.29

合衆

0.29

Activations Density 0.046%

instances and references related to ethical principles and frameworks

No Comments

No Known Activations