INDEX

Explanations

phrases related to political debates, winners, and personalities

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

1:0.04

2:0.08

3:0.06

4:0.07

5:0.10

6:0.03

7:0.07

8:0.10

9:0.13

10:0.13

11:0.06

Negative Logits

fecture

-1.09

ertodd

-1.03

 Units

-0.99

 pairs

-0.97

untarily

-0.93

 optionally

-0.93

龍

-0.92

apixel

-0.91

 subdivision

-0.91

 authorised

-0.90

POSITIVE LOGITS

 cynicism

1.09

gem

1.04

 mockery

1.00

 bigotry

1.00

 stunts

0.97

pun

0.95

 patriotism

0.95

bole

0.94

 tales

0.92

 satir

0.90

Activations Density 0.830%

phrases related to political debates, winners, and personalities

No Comments

No Known Activations