INDEX

Explanations

occurrences of object-oriented programming terms and structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_7/width_16k/average_l0_92

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 feroit

-0.60

 faſt

-0.57

anasia

-0.56

canActivate

-0.56

 ſhall

-0.56

validates

-0.56

caret

-0.55

 pleaſure

-0.54

 ſhould

-0.54

Fall

-0.54

POSITIVE LOGITS

obj

2.64

obj

2.55

Obj

2.09

Obj

1.90

OBJ

1.70

OBJ

1.59

objs

1.49

 objs

1.36

 objective

1.30

 Objek

1.23

Activations Density 0.049%

occurrences of object-oriented programming terms and structure

No Comments

No Known Activations