INDEX

Explanations

conditional statements and their associated logical structures in the text

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_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

0:0.03

1:0.03

2:0.07

3:0.17

4:0.09

5:0.05

6:0.20

7:0.03

8:0.04

9:0.07

10:0.09

11:0.08

Negative Logits

ARB

-1.59

 corrid

-1.35

ModLoader

-1.33

 behavi

-1.32

VID

-1.32

 unden

-1.30

ADRA

-1.29

senal

-1.28

[/

-1.20

ンジ

-1.19

POSITIVE LOGITS

unic

1.46

eman

1.34

ffect

1.34

urst

1.32

eem

1.32

unct

1.31

ammed

1.30

ensitive

1.30

amb

1.29

gn

1.29

Activations Density 0.003%

conditional statements and their associated logical structures in the text

No Comments

No Known Activations