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

0:0.08

1:0.09

2:0.09

3:0.08

4:0.09

5:0.07

6:0.07

7:0.08

8:0.08

9:0.06

10:0.08

11:0.08

Negative Logits

モ

-1.73

オ

-1.71

ヴァ

-1.58

abouts

-1.55

ラン

-1.51

 highways

-1.50

 topics

-1.47

 administered

-1.44

将

-1.43

っ

-1.37

POSITIVE LOGITS

roth

1.98

schild

1.94

blance

1.77

llor

1.77

rators

1.75

sonian

1.75

elo

1.72

ylum

1.69

umers

1.64

oulos

1.63

Activations Density 0.000%

No Known Activations

This feature has no known activations.