INDEX

Explanations

indicators of confidentiality or secrecy

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

New Auto-Interp

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

0:0.06

1:0.06

2:0.07

3:0.08

4:0.09

5:0.07

6:0.08

7:0.07

8:0.07

9:0.09

10:0.11

11:0.09

Negative Logits

terms

-1.64

zl

-1.54

ē

-1.45

®

-1.44

 Deity

-1.38

Desk

-1.37

otle

-1.35

}}

-1.33

 Lite

-1.29

 Runes

-1.28

POSITIVE LOGITS

kered

1.71

 clut

1.59

 populated

1.53

lict

1.45

licts

1.43

 Railway

1.41

 dysfunctional

1.39

?????-

1.37

 workforce

1.35

 epidem

1.34

Activations Density 0.000%

indicators of confidentiality or secrecy

No Comments

No Known Activations