INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Configuration

ckkissane/attn-saes-gpt2-small-all-layers/gpt2-small_L9_Hcat_z_lr1.20e-03_l11.20e+00_ds24576_bs4096_dc1.00e-06_rsanthropic_rie25000_nr4_v9.pt

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

float32

Hook Name

blocks.9.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

1:0.09

2:0.07

3:0.08

4:0.09

5:0.08

6:0.07

7:0.08

8:0.06

9:0.08

10:0.09

11:0.08

Negative Logits

 Slack

-3.04

 LEDs

-2.71

 responders

-2.61

 Lego

-2.54

OD

-2.53

 logger

-2.50

 McDonnell

-2.45

 Brigade

-2.45

renches

-2.40

 stranded

-2.37

POSITIVE LOGITS

aurus

3.02

ebus

2.79

pec

2.75

ape

2.73

uve

2.71

annis

2.56

Pau

2.53

peat

2.53

heit

2.50

orc

2.49

Activations Density 0.000%

No Known Activations

This feature has no known activations.