INDEX

Explanations

themes related to human flaws and societal critiques

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_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.05

1:0.02

2:0.07

3:0.24

4:0.02

5:0.05

6:0.02

7:0.04

8:0.02

9:0.01

10:0.38

11:0.03

Negative Logits

 similarities

-2.63

 essentials

-2.35

 strengths

-2.27

 significance

-2.20

 similarity

-2.16

CLUD

-2.10

 advantages

-2.08

 importance

-2.08

 dimensions

-2.04

Hig

-2.03

POSITIVE LOGITS

 tempted

2.43

 unwilling

2.38

 ignor

2.34

 gull

2.33

 careless

2.29

 misguided

2.29

 disob

2.28

 blindly

2.27

ardy

2.25

 ignorant

2.25

Activations Density 0.379%

themes related to human flaws and societal critiques

No Comments

No Known Activations