INDEX

Explanations

phrases indicating activities and actions related to organized groups or entities

New Auto-Interp

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

Head Attr Weights

0:0.01

1:0.01

2:0.08

3:0.04

4:0.10

5:0.01

6:0.04

7:0.40

8:0.03

9:0.03

10:0.12

11:0.08

Negative Logits

vals

-1.55

 slammed

-1.52

 rejected

-1.48

atts

-1.46

ropy

-1.43

faces

-1.42

 defeats

-1.42

 faces

-1.42

 shudder

-1.39

 faded

-1.38

POSITIVE LOGITS

 activities

1.91

meet

1.78

Interstitial

1.75

ecycle

1.64

 tourism

1.63

 recreation

1.63

 diplomacy

1.61

 fundraising

1.61

Reviewer

1.61

 Activities

1.60

Activations Density 0.007%

phrases indicating activities and actions related to organized groups or entities

No Known Activations