INDEX

Explanations

references to specific names or labels within a contextual framework

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

 Administración

-0.28

 gestellt

-0.24

 Gestión

-0.23

 הס

-0.21

<strong>

-0.21

mę

-0.21

‘

-0.20

</em>

-0.20

<eos>

-0.20

POSITIVE LOGITS

 للاسماء

0.92

 للمعارف

0.88

StoryboardSegue

0.84

TagMode

0.83

RenderAtEndOf

0.82

 المعيارى

0.81

 transfieras

0.79

 ✭✭

0.78

 simplifié

0.78

 TestBed

0.77

Activations Density 0.269%

references to specific names or labels within a contextual framework

No Comments

No Known Activations