INDEX

Explanations

No Explanations Found

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.08

1:0.06

2:0.08

3:0.10

4:0.07

5:0.08

6:0.07

7:0.07

8:0.09

9:0.08

10:0.09

11:0.08

Negative Logits

"],"

-1.59

 Maxim

-1.58

 Publisher

-1.55

FF

-1.51

 Units

-1.50

)].

-1.49

 Sell

-1.47

 Sold

-1.47

 Remastered

-1.45

 Print

-1.44

POSITIVE LOGITS

 slippery

1.65

orgetown

1.64

 regress

1.62

 sucker

1.58

 suspicious

1.56

 trusting

1.55

 irony

1.54

 rude

1.54

 cuff

1.52

 seeming

1.51

Activations Density 0.000%

No Known Activations

This feature has no known activations.