INDEX

Explanations

references to publications and articles, especially those related to academic and research work

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

0:0.16

1:0.02

2:0.05

3:0.13

4:0.08

5:0.14

6:0.03

7:0.09

8:0.07

9:0.02

10:0.11

11:0.03

Negative Logits

rators

-1.96

 actor

-1.88

 impe

-1.86

police

-1.86

 pacif

-1.72

 inspecting

-1.71

 disband

-1.68

SEA

-1.66

rator

-1.65

 patriotism

-1.65

POSITIVE LOGITS

doi

2.87

Abstract

2.79

DOI

2.73

 Abstract

2.66

 Neuroscience

2.59

 Clinical

2.54

 Xiang

2.44

Fei

2.40

doi

2.30

 neuroscience

2.27

Activations Density 0.010%

references to publications and articles, especially those related to academic and research work

No Comments

No Known Activations