INDEX

Explanations

phrases related to accidents and safety concerns involving firearms

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

0:0.11

1:0.03

2:0.21

3:0.07

4:0.14

5:0.05

6:0.06

7:0.05

8:0.08

9:0.05

10:0.05

11:0.05

Negative Logits

orem

-1.82

ico

-1.71

oris

-1.58

️

-1.58

eness

-1.55

orio

-1.52

oso

-1.51

resa

-1.49

orie

-1.49

aha

-1.48

POSITIVE LOGITS

 Nether

1.66

berus

1.65

 Levant

1.50

 Reincarn

1.44

 Extras

1.44

aples

1.41

roleum

1.35

 guiActiveUn

1.32

dfx

1.27

 skelet

1.26

Activations Density 0.001%

phrases related to accidents and safety concerns involving firearms

No Comments

No Known Activations