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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.09

1:0.09

2:0.07

3:0.07

4:0.09

5:0.09

6:0.08

7:0.07

8:0.09

9:0.08

10:0.06

11:0.08

Negative Logits

SQL

-1.88

�

-1.77

Setting

-1.76

record

-1.72

dict

-1.71

thora

-1.71

�

-1.68

pec

-1.68

�

-1.60

DEV

-1.59

POSITIVE LOGITS

 promoters

1.68

 introdu

1.63

 dodging

1.59

 Karin

1.57

 spokes

1.56

 glared

1.55

 flanked

1.55

 barr

1.54

oon

1.53

 doub

1.52

Activations Density 0.000%

No Known Activations

This feature has no known activations.