INDEX

Explanations

mentions of scientific journals and publication references

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

0:0.06

1:0.01

2:0.11

3:0.34

4:0.02

5:0.11

6:0.02

7:0.04

8:0.07

9:0.03

10:0.08

11:0.05

Negative Logits

TRUMP

-2.53

 hashtag

-2.20

 flag

-2.17

 banners

-2.17

 misplaced

-2.15

 swast

-2.12

 hypocrisy

-2.02

 TRUMP

-2.01

 brazen

-1.94

 hasht

-1.93

POSITIVE LOGITS

iatrics

2.99

 Proceedings

2.67

 Neuroscience

2.62

 Diseases

2.54

 Biology

2.49

sci

2.48

Studies

2.45

Psych

2.38

Phys

2.36

biology

2.34

Activations Density 0.051%

mentions of scientific journals and publication references

No Comments

No Known Activations