INDEX

Explanations

discussions or arguments that involve opinions and beliefs

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

personal pronouns and associated actions or states, particularly focusing on individual perspectives and responsibilities

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

Not in Any Lists

No Comments

Head Attr Weights

0:0.02

1:0.09

2:0.11

3:0.05

4:0.01

5:0.06

6:0.06

7:0.06

8:0.13

9:0.15

10:0.06

11:0.14

Negative Logits

 Bucket

-1.04

 Rewards

-1.01

NBA

-1.01

Pacific

-1.00

 Starting

-0.98

Ball

-0.96

bowl

-0.96

RGB

-0.96

 Balloon

-0.95

tick

-0.95

POSITIVE LOGITS

 anymore

1.95

nor

1.86

 necessarily

1.33

 altogether

1.31

 slightest

1.28

any

1.25

pse

1.23

 thereof

1.21

 remotely

1.18

 milit

1.18

Activations Density 0.018%

discussions or arguments that involve opinions and beliefs

personal pronouns and associated actions or states, particularly focusing on individual perspectives and responsibilities

No Comments

No Known Activations