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

0:0.06

1:0.05

2:0.07

3:0.09

4:0.08

5:0.08

6:0.08

7:0.09

8:0.08

9:0.10

10:0.09

11:0.09

Negative Logits

AGES

-1.93

GGGGGGGG

-1.84

Los

-1.83

ONSORED

-1.69

 Beir

-1.69

ITE

-1.65

ItemImage

-1.60

"],"

-1.57

ALE

-1.54

rored

-1.53

POSITIVE LOGITS

 hindsight

1.91

 retrospect

1.64

fix

1.58

lua

1.50

 regenerate

1.47

ql

1.47

 libel

1.45

 instant

1.45

 impeachment

1.42

 cheat

1.42

Activations Density 0.000%

No Known Activations

This feature has no known activations.