INDEX

Explanations

definitions and discussions about classification and criteria

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

2:0.29

3:0.11

4:0.04

5:0.06

6:0.03

7:0.06

8:0.04

9:0.04

10:0.05

11:0.04

Negative Logits

 Roose

-3.23

RIP

-2.68

Haunted

-2.55

 Vict

-2.52

warming

-2.50

SIGN

-2.44

Rew

-2.41

 rematch

-2.39

timeout

-2.37

 waited

-2.37

POSITIVE LOGITS

 definitions

5.78

 definition

5.47

 distinctions

5.32

definition

5.09

 classification

4.93

 terminology

4.85

 labels

4.74

 Definition

4.70

 categories

4.66

 distinction

4.57

Activations Density 0.375%

definitions and discussions about classification and criteria

No Comments

No Known Activations