INDEX

Explanations

meta-information related to sources and references

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

ungeons

-0.53

Identyfik

-0.52

انيف

-0.52

 túl

-0.52

 VIDEOTAPE

-0.51

 pousser

-0.50

 pytest

-0.49

 crochê

-0.49

 sonno

-0.48

 esetén

-0.48

POSITIVE LOGITS

 تضيفلها

0.63

ImageContext

0.62

 unofficial

0.60

ConstraintMaker

0.59

########.

0.58

 moul

0.51

Personendaten

0.51

 AspNetCore

0.51

swire

0.51

 Exacts

0.50

Activations Density 0.033%

meta-information related to sources and references

No Comments

No Known Activations

meta-information related to sources and references

No Comments

No Known Activations