INDEX

Explanations

connections and interactions within various fields and industries

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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.04

1:0.04

2:0.05

3:0.25

4:0.02

5:0.04

6:0.02

7:0.04

8:0.03

9:0.02

10:0.37

11:0.03

Negative Logits

 adjusting

-1.91

 Bender

-1.70

 rest

-1.68

 warmer

-1.68

 trying

-1.65

 separating

-1.65

 recovering

-1.59

 tonight

-1.58

 contrace

-1.57

 Nathan

-1.57

POSITIVE LOGITS

thinkable

2.85

stros

2.39

ceivable

2.31

Such

2.27

idable

2.23

 suffice

2.19

ocious

2.14

 ausp

2.13

quickShipAvailable

2.10

orthy

2.08

Activations Density 0.070%

connections and interactions within various fields and industries

No Comments

No Known Activations