INDEX

Explanations

references to the concept of 'view' or 'profile'

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

CardContent

-0.79

AlterField

-0.76

fontawesome

-0.74

 réglementation

-0.73

expandindo

-0.72

kowym

-0.72

 Arkivert

-0.71

 BorderRadius

-0.71

 Geheimnis

-0.71

 Bracken

-0.70

POSITIVE LOGITS

 VIEW

1.88

 view

1.88

 views

1.86

 View

1.75

 Views

1.64

view

1.62

 VIEWS

1.61

VIEW

1.57

views

1.55

View

1.46

Activations Density 0.083%

references to the concept of 'view' or 'profile'

No Comments

No Known Activations