INDEX

Explanations

technical language related to materials or products and advancements in technology

oai_token-act-pair · gpt-3.5-turbo

Triggered by @bot

terms and phrases associated with improvement or enhancement in various contexts

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-att-kk

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Head Attr Weights

0:0.01

1:0.03

2:0.08

3:0.14

4:0.01

5:0.04

6:0.05

7:0.10

8:0.09

9:0.14

10:0.07

11:0.16

Negative Logits

INGTON

-1.21

lers

-1.08

archive

-1.04

Haunted

-1.04

issors

-1.04

EStream

-1.01

channelAvailability

-1.00

glers

-0.99

tails

-0.99

BACK

-0.99

POSITIVE LOGITS

 capabilities

1.39

than

1.32

 functionality

1.21

 image

1.18

 capability

1.17

amina

1.15

 exponentially

1.13

 visuals

1.12

 visibility

1.11

 abilities

1.10

Activations Density 0.016%

technical language related to materials or products and advancements in technology

terms and phrases associated with improvement or enhancement in various contexts

No Known Activations