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

0:0.05

1:0.05

2:0.09

3:0.09

4:0.08

5:0.07

6:0.08

7:0.10

8:0.08

9:0.09

10:0.09

11:0.08

Negative Logits

 altogether

-1.52

 himself

-1.48

 ransom

-1.43

 anymore

-1.42

heter

-1.38

utsch

-1.37

content

-1.34

 schizophren

-1.34

 disproportion

-1.33

....

-1.33

POSITIVE LOGITS

 Cutter

1.64

inventoryQuantity

1.56

lasses

1.55

quickShipAvailable

1.53

verbs

1.53

TYPE

1.53

athi

1.52

ELD

1.51

zees

1.50

 Walking

1.48

Activations Density 0.000%

No Known Activations

This feature has no known activations.