INDEX

Explanations

instances of increases or changes in quantitative data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 40-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_40/width_16k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.40.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

 CreateTagHelper

-0.65

 للاسماء

-0.55

OCCURRED

-0.52

AndEndTag

-0.47

ammans

-0.45

новништво

-0.45

NameInMap

-0.45

 szóci

-0.43

jarkan

-0.42

 mengin

-0.42

POSITIVE LOGITS

 spike

1.05

 surge

0.98

 boost

0.97

 rise

0.97

 increase

0.96

 shift

0.94

 jump

0.92

 decrease

0.85

 drop

0.85

 bump

0.85

Activations Density 0.671%

instances of increases or changes in quantitative data

No Comments

No Known Activations