INDEX

Explanations

actions involving falling, tripping, or being obstructed

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

Not in Any Lists

No Comments

Head Attr Weights

0:0.02

1:0.01

2:0.07

3:0.06

4:0.12

5:0.03

6:0.11

7:0.36

8:0.02

9:0.04

10:0.06

11:0.05

Negative Logits

CLUS

-1.83

sonian

-1.82

)=(

-1.67

shown

-1.56

PDATE

-1.53

女

-1.49

displayText

-1.49

ヴ

-1.47

 congr

-1.45

eta

-1.44

POSITIVE LOGITS

 overhe

1.95

 stumbling

1.65

 grappling

1.59

 shaky

1.57

 grasp

1.56

 rocky

1.46

 choking

1.45

 ropes

1.45

 hitch

1.44

 bells

1.42

Activations Density 0.001%

actions involving falling, tripping, or being obstructed

No Comments

No Known Activations