INDEX

Explanations

references to specific events or entities related to American aviation history

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

references to specific countries and organizations related to the Pan-American context

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

0:0.03

1:0.04

2:0.13

3:0.10

4:0.03

5:0.02

6:0.16

7:0.09

8:0.04

9:0.09

10:0.07

11:0.14

Negative Logits

 grievances

-1.00

iott

-1.00

ibles

-0.98

 CLASS

-0.98

 Vance

-0.94

 Oath

-0.94

 wisely

-0.94

td

-0.94

 Meredith

-0.92

___

-0.92

POSITIVE LOGITS

 Flavoring

1.25

adan

1.13

ayan

1.06

iterranean

1.02

uters

1.01

apan

0.99

ophob

0.99

ascus

0.98

acht

0.97

ゼウス

0.96

Activations Density 0.006%

references to specific events or entities related to American aviation history

references to specific countries and organizations related to the Pan-American context

No Comments

No Known Activations