INDEX

Explanations

references to authoritative sources or frameworks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

enegro

-0.52

 bide

-0.47

autoload

-0.44

 JSTOR

-0.43

figcaption

-0.43

vière

-0.43

NewUrlParser

-0.43

usda

-0.42

DatabaseError

-0.41

 стаття

-0.40

POSITIVE LOGITS

 Consistency

0.90

 uniformity

0.88

 inconsistent

0.86

 consistency

0.85

consistent

0.82

consistency

0.81

統一

0.79

Consistency

0.79

 inconsistency

0.79

 inconsistencies

0.77

Activations Density 0.356%

references to authoritative sources or frameworks

No Comments

No Known Activations