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

1:0.06

2:0.08

3:0.08

4:0.09

5:0.08

6:0.09

7:0.08

8:0.07

9:0.08

10:0.09

11:0.08

Negative Logits

resent

-1.78

 equals

-1.67

ament

-1.60

 legisl

-1.56

aments

-1.52

 Constant

-1.51

 Fiat

-1.50

igil

-1.50

 reserv

-1.47

 volunt

-1.47

POSITIVE LOGITS

fect

1.89

 VIDEOS

1.78

clips

1.70

 sucker

1.70

girls

1.59

 snapshots

1.59

 aeros

1.58

azard

1.56

thritis

1.54

 contam

1.53

Activations Density 0.000%

No Known Activations

This feature has no known activations.