INDEX

Explanations

expressions of opinion or judgment about various topics

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

0:0.04

1:0.03

2:0.05

3:0.14

4:0.03

5:0.06

6:0.03

7:0.05

8:0.03

9:0.01

10:0.45

11:0.02

Negative Logits

"]=>

-2.33

 prevailed

-2.09

 overturned

-2.04

 upheld

-2.04

 substituted

-2.03

 besieged

-1.98

 postponed

-1.98

 "$:/

-1.97

anwhile

-1.97

 vacated

-1.95

POSITIVE LOGITS

inav

2.10

vell

2.05

acid

2.03

esi

2.01

acist

1.92

sav

1.89

��

1.87

urus

1.87

catentry

1.87

igr

1.86

Activations Density 0.008%

expressions of opinion or judgment about various topics

No Comments

No Known Activations