INDEX

Explanations

concepts related to events, networking, and history

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

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.3.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.02

1:0.11

2:0.18

3:0.04

4:0.01

5:0.02

6:0.10

7:0.26

8:0.13

9:0.03

10:0.03

11:0.03

Negative Logits

────

-1.40

 Grail

-1.27

VERTISEMENT

-1.17

\\\\\\\\

-1.11

aste

-1.08

ussions

-1.07

 Shroud

-1.04

 sorts

-1.01

 redund

-1.00

rogens

-1.00

POSITIVE LOGITS

pport

1.39

azeera

1.35

omal

1.35

iaries

1.27

Downloadha

1.27

uper

1.26

ocr

1.23

ensical

1.15

eric

1.15

cience

1.14

Activations Density 0.558%

concepts related to events, networking, and history

No Comments

No Known Activations