INDEX

Explanations

No Explanations Found

New Auto-Interp

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.07

1:0.08

2:0.07

3:0.06

4:0.08

5:0.09

6:0.08

7:0.09

8:0.08

9:0.08

10:0.08

11:0.08

Negative Logits

Newsletter

-1.78

[|

-1.77

Û

-1.62

-1.61

 Airbus

-1.58

 veter

-1.53

APD

-1.53

earch

-1.52

orea

-1.44

OTA

-1.43

POSITIVE LOGITS

 snapped

1.55

 fists

1.41

 Tears

1.40

 cred

1.35

 specimen

1.35

 endured

1.35

 leverage

1.34

 shoulders

1.34

 face

1.33

 instinct

1.32

Activations Density 0.000%

No Known Activations

This feature has no known activations.