INDEX

Explanations

terms related to academic institutions and their locations

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_L5_Hcat_z_lr1.20e-03_l11.00e+00_ds49152_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

float32

Hook Name

blocks.5.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

0:0.01

1:0.01

2:0.04

3:0.05

4:0.03

5:0.03

6:0.15

7:0.35

8:0.03

9:0.05

10:0.12

11:0.07

Negative Logits

 ration

-1.79

 aisle

-1.70

byn

-1.69

 pedest

-1.62

 bandwagon

-1.58

 topp

-1.58

 coffin

-1.58

 litter

-1.55

 choices

-1.52

 buttons

-1.48

POSITIVE LOGITS

 Sciences

1.79

anu

1.78

 Robotics

1.74

 Uzbek

1.72

 Anthropology

1.69

uador

1.67

 Mend

1.65

 Hebrew

1.62

Math

1.59

 Prague

1.59

Activations Density 0.001%

terms related to academic institutions and their locations

No Comments

No Known Activations