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

0:0.08

1:0.07

2:0.09

3:0.09

4:0.08

5:0.08

6:0.08

7:0.08

8:0.07

9:0.07

10:0.07

11:0.08

Negative Logits

 Bulgar

-1.88

Brit

-1.78

 Cavs

-1.78

 Reincarn

-1.75

 Beet

-1.72

 Wolver

-1.69

UCH

-1.65

 Brit

-1.63

 Scots

-1.63

 Hitch

-1.62

POSITIVE LOGITS

ilde

1.88

isoft

1.87

urance

1.86

inventoryQuantity

1.70

livious

1.69

 hindsight

1.66

iversity

1.65

mberg

1.65

conservancy

1.63

management

1.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.