INDEX

Explanations

timestamps and dates in the text

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

0:0.01

1:0.01

2:0.13

3:0.16

4:0.05

5:0.01

6:0.04

7:0.05

8:0.05

9:0.04

10:0.27

11:0.14

Negative Logits

 suites

-1.55

 suite

-1.54

 waivers

-1.53

 Gloss

-1.44

 Advantage

-1.44

 strengths

-1.43

 envy

-1.42

 coefficients

-1.41

 lect

-1.41

 Exam

-1.39

POSITIVE LOGITS

 millenn

1.66

 uploaded

1.60

 Upload

1.60

iframe

1.57

onica

1.56

 uploading

1.54

imedia

1.53

malink

1.53

imen

1.48

 Tags

1.47

Activations Density 0.006%

timestamps and dates in the text

No Comments

No Known Activations