INDEX

Explanations

content related to incidents involving discussions or clarifications surrounding significant events or cases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

styleType

-0.66



-0.61

 Anſ

-0.59

目指

-0.57

новништво

-0.56

TagMode

-0.56

 pleaſure

-0.54

 referenties

-0.53

envie

-0.53

 hâte

-0.52

POSITIVE LOGITS

 incident

2.27

 incidents

1.82

incident

1.75

 Incident

1.68

 incidente

1.53

事件

1.48

 events

1.42

 Incidents

1.42

 kejadian

1.35

Incident

1.35

Activations Density 0.676%

content related to incidents involving discussions or clarifications surrounding significant events or cases

No Comments

No Known Activations