INDEX

Explanations

structures or symbols commonly used in programming or data processing

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_L10_Hcat_z_lr1.20e-03_l11.30e+00_ds24576_bs4096_dc1.00e-05_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.10.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

0:0.13

1:0.07

2:0.08

3:0.06

4:0.11

5:0.05

6:0.04

7:0.15

8:0.09

9:0.03

10:0.08

11:0.07

Negative Logits

idium

-2.40

 arte

-2.36

mor

-2.33

 skelet

-2.31

 concess

-2.29

 divers

-2.28

 Osiris

-2.27

 mosa

-2.27

 Nile

-2.27

 skew

-2.25

POSITIVE LOGITS

____

4.49

_____

3.98

________

3.98

______

3.77

________________________________

3.61

___

3.56

________________________

3.46

_______

3.44

 ______

3.40

________________________________________________________________

3.27

Activations Density 0.008%

structures or symbols commonly used in programming or data processing

No Comments

No Known Activations