INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

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

0:0.07

1:0.07

2:0.08

3:0.08

4:0.08

5:0.09

6:0.09

7:0.08

8:0.08

9:0.08

10:0.08

11:0.06

Negative Logits

 invalid

-1.64

 discrepancies

-1.63

 suspicions

-1.57

 disagreement

-1.55

 argument

-1.55

Shares

-1.52

JP

-1.51

 retweet

-1.50

 hears

-1.49

 Heidi

-1.48

POSITIVE LOGITS

 tremend

2.09

aceutical

1.98

apons

1.84

aeda

1.83

�

1.72

 foremost

1.70

thood

1.69

 sparing

1.65

astered

1.65

 civilisation

1.61

Activations Density 0.000%

No Known Activations

This feature has no known activations.