INDEX

Explanations

content related to criticism and evaluation of beliefs or institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

 endblock

-0.43

EndContext

-0.43

Bronnen

-0.43

 Pekan

-0.42

MessageTagHelper

-0.40

 pinulongan

-0.38

DispatchToProps

-0.37

 Thane

-0.37

pê

-0.36

guenos

-0.36

POSITIVE LOGITS

 crap

1.82

 nonsense

1.70

 bullshit

1.66

 rubbish

1.61

 garbage

1.52

 shit

1.48

BS

1.38

 stuff

1.38

crap

1.34

nonsense

1.32

Activations Density 0.493%

content related to criticism and evaluation of beliefs or institutions

No Comments

No Known Activations