INDEX

Explanations

phrases indicating progress or the passage of time

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_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

2:0.08

3:0.08

4:0.02

5:0.04

6:0.08

7:0.16

8:0.12

9:0.05

10:0.08

11:0.20

Negative Logits

 immediately

-1.24

irs

-1.16

beer

-1.14

 instantly

-1.09

 Immediately

-1.09

properties

-1.03

fecture

-1.02

must

-1.01

dl

-1.00

aria

-0.99

POSITIVE LOGITS

 adolesc

1.22

 stint

1.14

 fulfilling

1.11

 anecd

1.08

VALUE

1.05

 charact

1.05

 millenn

1.04

 enthusi

1.04

ihara

1.04

 unexplained

1.03

Activations Density 0.012%

phrases indicating progress or the passage of time

No Comments

No Known Activations