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

2:0.07

3:0.08

4:0.07

5:0.09

6:0.09

7:0.10

8:0.08

9:0.07

10:0.08

11:0.08

Negative Logits

RAW

-1.95

powder

-1.86

perial

-1.85

anamo

-1.79

atts

-1.77

area

-1.72

cest

-1.72

raped

-1.68

ascus

-1.68

bur

-1.67

POSITIVE LOGITS

 minded

1.77

 Raphael

1.65

 whoever

1.62

 motivating

1.58

 biased

1.54

 Gupta

1.50

 Gabriel

1.48

 Boolean

1.47

 limiting

1.47

irements

1.47

Activations Density 0.000%

No Known Activations

This feature has no known activations.