INDEX

Explanations

affirmative action phrases indicating movement or progression

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_L7_Hcat_z_lr1.20e-03_l11.10e+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.7.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

0:0.08

1:0.08

2:0.06

3:0.08

4:0.08

5:0.09

6:0.08

7:0.06

8:0.08

9:0.08

10:0.08

11:0.09

Negative Logits

acid

-2.52

 Nicarag

-2.33

terrorist

-2.24

eteria

-2.23

iferation

-2.22

unal

-2.19

mire

-2.14

iable

-2.13

osion

-2.13

ophobic

-2.13

POSITIVE LOGITS

 merch

2.10

Pil

2.09

 Including

2.06

 comprising

2.06

 Merch

2.05

 Serving

2.02

 Revenue

2.01

 commissions

2.00

 Shipping

2.00

 Mess

1.99

Activations Density 0.000%

affirmative action phrases indicating movement or progression

No Comments

No Known Activations