INDEX

Explanations

references to visiting or accessing information, particularly through websites or email

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

1:0.02

2:0.06

3:0.14

4:0.31

5:0.02

6:0.05

7:0.07

8:0.04

9:0.05

10:0.09

11:0.07

Negative Logits

 sucked

-1.92

 perceived

-1.86

 creeping

-1.68

 suddenly

-1.66

 lied

-1.66

 prey

-1.59

 falsely

-1.56

 passers

-1.55

 mentally

-1.54

 distorted

-1.53

POSITIVE LOGITS

Sources

1.87

 Archives

1.78

forum

1.78

pedia

1.76

 Kamp

1.75

Camp

1.69

FAQ

1.68

Ak

1.65

 archives

1.64

Rh

1.61

Activations Density 0.088%

references to visiting or accessing information, particularly through websites or email

No Comments

No Known Activations