INDEX

Explanations

code constructs related to methods and control flow in programming

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

rugada

-0.50

睫

-0.49

-0.47

աք

-0.47

Par

-0.47

\}\\

-0.46

libft

-0.46

Frame

-0.46

 Frame

-0.46

Qu

-0.45

POSITIVE LOGITS

 myſelf

0.73

 InputDecoration

0.69

ArgumentParser

0.66

 Chriſt

0.61

 ſhe

0.61

 himſelf

0.60

 Monfieur

0.60

msgTypes

0.59

 themſelves

0.58

Efq

0.55

Activations Density 0.233%

code constructs related to methods and control flow in programming

No Comments

No Known Activations