INDEX

Explanations

emphatic qualifiers indicating uniqueness or exclusivity

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

0:0.07

1:0.07

2:0.09

3:0.08

4:0.07

5:0.08

6:0.08

7:0.08

8:0.09

9:0.09

10:0.07

11:0.08

Negative Logits

Iowa

-2.94

 caucuses

-2.93

interstitial

-2.89

asus

-2.86

enary

-2.83

 princ

-2.73

pox

-2.66

Corn

-2.60

sow

-2.60

BUS

-2.57

POSITIVE LOGITS

 Barcl

3.48

 Gear

3.16

 Destination

2.84

 Zimmer

2.79

 Morrow

2.77

 Deck

2.74

 Torrent

2.73

 Merchant

2.72

 Chrom

2.72

 Outer

2.65

Activations Density 0.000%

emphatic qualifiers indicating uniqueness or exclusivity

No Comments

No Known Activations