INDEX

Explanations

references to various incidents

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 Carthage

-0.70

 marito

-0.64

rmdir

-0.63

thag

-0.61

tidad

-0.58

JMenu

-0.58

 CURLOPT

-0.57

 Farrell

-0.56

 cesta

-0.56

rawValue

-0.55

POSITIVE LOGITS

 incident

2.57

 Incident

2.37

 incidents

2.27

incident

2.25

 Incidents

1.97

Incident

1.86

 incidente

1.47

 incidence

1.40

 Incidence

1.37

 incid

1.28

Activations Density 0.080%

references to various incidents

No Comments

No Known Activations

references to various incidents

No Comments

No Known Activations