INDEX

Explanations

expressions of uncertainty or doubt

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

expressions of personal capabilities and emotions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-att-kk

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.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

0:0.01

1:0.03

2:0.11

3:0.14

4:0.01

5:0.03

6:0.04

7:0.16

8:0.19

9:0.08

10:0.06

11:0.08

Negative Logits

ARB

-1.19

istries

-1.09

mt

-0.98

abled

-0.95

 citing

-0.94

REE

-0.94

azeera

-0.93

incial

-0.93

oggle

-0.92

NJ

-0.91

POSITIVE LOGITS

 underestimated

1.26

 comprom

1.14

 Subtle

1.10

 Puppet

1.09

 Plug

1.07

 Cyborg

1.03

hack

1.02

zin

1.01

 Photographer

1.00

lator

1.00

Activations Density 0.022%

expressions of uncertainty or doubt

expressions of personal capabilities and emotions

No Comments

No Known Activations