INDEX

Explanations

sections related to code documentation and method summaries

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

/**

-0.53

中略

-0.45

icol

-0.43

 varargin

-0.42

 espagnole

-0.42

 tomen

-0.41

 Waterford

-0.41

=>

-0.41

Willkommen

-0.41

orations

-0.41

POSITIVE LOGITS

 snippetHide

0.77

 ExecuteAsync

0.77

Datuak

0.77

 externi

0.74

expandindo

0.72

bootstrapcdn

0.72

 незавершена

0.69

AccessorTable

0.68

principalTable

0.68

AndEndTag

0.68

Activations Density 0.049%

sections related to code documentation and method summaries

No Comments

No Known Activations