INDEX

Explanations

code-related terms, particularly those associated with class and method definitions in programming languages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.13.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

 Shakspeare

-0.85

 Anſ

-0.84

 purpoſe

-0.83

 Houſe

-0.83

 photolibrary

-0.82

ergies

-0.79

 étoient

-0.77

 avoient

-0.77

 Roskov

-0.76

 myſelf

-0.76

POSITIVE LOGITS

GOTREF

0.59

“

0.56

0.55

0.54

Inc

0.51

0.48

tania

0.46

としても

0.46

Activations Density 0.140%

code-related terms, particularly those associated with class and method definitions in programming languages

No Comments

No Known Activations