INDEX

Explanations

terms related to monitoring and supervision

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

GeneratedValue

-0.66

-0.61

 Cous

-0.60

vä

-0.59

Де

-0.57

cinia

-0.57

澡

-0.56

la

-0.55

まさ

-0.55

POSITIVE LOGITS

 monitors

2.48

 monitor

2.42

 monitoring

2.30

 Monitors

2.28

 Monitor

2.23

 MONITOR

2.21

 Monitoring

2.17

MONITOR

2.17

monitor

2.14

Monitor

2.13

Activations Density 0.063%

terms related to monitoring and supervision

No Comments

No Known Activations