INDEX

Explanations

guidance on skills development and improvement through practice

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_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.9.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

0:0.16

1:0.02

2:0.28

3:0.12

4:0.03

5:0.09

6:0.03

7:0.07

8:0.04

9:0.03

10:0.06

11:0.02

Negative Logits

 crem

-2.74

soType

-2.64

 condol

-2.59

 UNHCR

-2.54

xit

-2.46

 protestors

-2.45

 vetoed

-2.38

 creditor

-2.38

estate

-2.36

 boycot

-2.34

POSITIVE LOGITS

 learners

5.64

 learning

5.64

Learning

5.33

learning

5.21

 lear

5.13

 Learning

5.07

 mastery

5.03

learn

4.96

 skills

4.91

 beginners

4.58

Activations Density 0.757%

guidance on skills development and improvement through practice

No Comments

No Known Activations