INDEX

Explanations

numbers or lists within a given document

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.02

2:0.10

3:0.09

4:0.04

5:0.14

6:0.04

7:0.02

8:0.16

9:0.15

10:0.08

11:0.04

Negative Logits

vironment

-1.25

byss

-1.24

ォ

-1.21

��

-1.18

ワン

-1.11

��

-1.04

akings

-1.03

idity

-1.03

 Dani

-1.01

vous

-1.01

POSITIVE LOGITS

ebin

1.07

 fundra

1.06

bugs

0.98

helps

0.93

ecause

0.92

want

0.92

 synerg

0.91

hower

0.91

 barg

0.90

 folklore

0.90

Activations Density 0.105%

numbers or lists within a given document

No Comments

No Known Activations