INDEX

Explanations

errors in code 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

0:0.07

1:0.05

2:0.09

3:0.12

4:0.07

5:0.14

6:0.07

7:0.06

8:0.07

9:0.08

10:0.08

11:0.04

Negative Logits

 Canaver

-1.26

PHOTOS

-1.17

pmwiki

-1.09

watching

-1.09

uploads

-1.07

views

-1.07

reports

-1.05

 Dahl

-1.05

 VIDEOS

-1.05

 confir

-1.04

POSITIVE LOGITS

$.

1.23

hai

1.18

regate

1.17

iman

1.14

wan

1.10

Chan

1.08

$,

1.05

Insert

1.05

Applic

1.03

lesi

1.03

Activations Density 0.006%

errors in code statements

No Comments

No Known Activations