INDEX

Explanations

instances of potential actions or recommendations made in a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 Roskov

-0.40

templateUrl

-0.37

 Савезне

-0.35

 transfieras

-0.35

tagHelperRunner

-0.35

 său

-0.33

)}_

-0.33

 franqu

-0.33

 certified

-0.32

onade

-0.31

POSITIVE LOGITS

WebVitals

0.56

 resourceCulture

0.53

 Infórmanos

0.52

RenderAtEndOf

0.49

IntoConstraints

0.47

 ModelRenderer

0.45

uxxxx

0.44

Peg

0.44

inversion

0.43

police

0.43

Activations Density 0.563%

instances of potential actions or recommendations made in a narrative context

No Comments

No Known Activations