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

0:0.06

1:0.08

2:0.08

3:0.08

4:0.08

5:0.08

6:0.08

7:0.08

8:0.09

9:0.07

10:0.08

11:0.08

Negative Logits

 deft

-1.84

 kindred

-1.78

urally

-1.73

 Quart

-1.71

antha

-1.69

 roundup

-1.67

 cuff

-1.66

 chores

-1.66

 reperto

-1.64

 paperback

-1.63

POSITIVE LOGITS

�

2.01

 Claim

1.95

Claim

1.86

�

1.77

�

1.76

�

1.72

人

1.72

enda

1.71

�

1.69

�

1.68

Activations Density 0.000%

No Known Activations

This feature has no known activations.