INDEX

Explanations

numerically significant data points or references

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

IFrame

Link

Negative Logits

 المعرف

-0.63

 CURIAM

-0.57

styleType

-0.56

orianCalendar

-0.55

RDP

-0.54

 cade

-0.54

GINIA

-0.52

первых

-0.51

thâu

-0.50

خصة

-0.50

POSITIVE LOGITS

TagMode

0.70

 SEDS

0.66

<tbody>

0.63

MigrationBuilder

0.60

 Pyrene

0.55

Jîn

0.54

SourceChecksum

0.54

colgroup

0.53

HideFlags

0.53

Искәрмәләр

0.53

Activations Density 0.002%

numerically significant data points or references

No Known Activations