INDEX

Explanations

names of specific individuals or organizations

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L2_Hcat_z_lr1.20e-03_l11.00e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v4.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.2.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

1:0.03

2:0.10

3:0.07

4:0.02

5:0.12

6:0.06

7:0.12

8:0.05

9:0.05

10:0.17

11:0.09

Negative Logits

istries

-0.80

 […]

-0.79

mine

-0.77

 Console

-0.77

.",

-0.74

."[

-0.74

rush

-0.72

".[

-0.72

'"

-0.71

 atmosphere

-0.70

POSITIVE LOGITS

ニ

0.99

PDATE

0.94

arently

0.94

=~=~

0.94

reditary

0.91

Narr

0.87

ngth

0.87

Byz

0.86

姫

0.85

ンジ

0.83

Activations Density 0.059%

names of specific individuals or organizations

No Comments

No Known Activations