INDEX

Explanations

expressions of negation or denial

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

KommentareTeilen

-0.74

 Muffins

-0.73

raszam

-0.72

 fibrillation

-0.71

MPC

-0.69

:]:

-0.69

UNEP

-0.69

aspectj

-0.69

StateToProps

-0.68

ilesh

-0.68

POSITIVE LOGITS

 Never

1.55

 NEVER

1.49

NEVER

1.48

 never

1.48

Never

1.46

never

1.41

 EVER

1.21

 Nunca

1.16

 Ever

1.11

Nunca

1.11

Activations Density 0.050%

expressions of negation or denial

No Comments

No Known Activations