INDEX

Explanations

programming-related structure and syntax

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

rep

-0.45

 Figures

-0.44

morgen

-0.41

ereq

-0.41

uez

-0.41

-0.40

Figures

-0.40

daten

-0.40

ewon

-0.39

inSlope

-0.39

POSITIVE LOGITS

########.

0.77

ArgsConstructor

0.61

InjectAttribute

0.61

ConverterFactory

0.57

AutoScaleMode

0.49

تقاوى

0.49

RTEE

0.46

AndEndTag

0.46

 himo

0.45

Spoljašnje

0.44

Activations Density 0.047%

programming-related structure and syntax

No Comments

No Known Activations