INDEX

Explanations

direct and assertive statements

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

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

1:0.02

2:0.05

3:0.10

4:0.04

5:0.12

6:0.03

7:0.07

8:0.16

9:0.09

10:0.13

11:0.06

Negative Logits

ć

-1.18

VK

-0.96

raq

-0.95

SPONSORED

-0.94

utra

-0.91

imilation

-0.89

hal

-0.86

 GOODMAN

-0.86

 behold

-0.85

 Maced

-0.85

POSITIVE LOGITS

stretched

0.94

 filler

0.92

EXP

0.90

 logistics

0.88

isco

0.86

 detail

0.81

 weekly

0.80

enegger

0.79

 Spur

0.79

 bragging

0.79

Activations Density 0.250%

direct and assertive statements

No Comments

No Known Activations