INDEX

Explanations

keywords and identifiers related to programming structure and definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 CreateTagHelper

-0.88

 Chwiliwch

-0.80

 Paglinawan

-0.78

AccessorTable

-0.76

 Geſch

-0.72

protoimpl

-0.69

EDEFAULT

-0.69

principalColumn

-0.68

enablog

-0.67

MigrationBuilder

-0.66

POSITIVE LOGITS

 dite

0.37

Sec

0.35

Sy

0.34

…

0.33

Fa

0.32

 Cloud

0.31

 Vort

0.31

 Plain

0.30

Bri

0.30

….

0.30

Activations Density 0.028%

keywords and identifiers related to programming structure and definitions

No Comments

No Known Activations