INDEX

Explanations

the word "as" used to indicate comparisons or explanations

oai_token-act-pair · gpt-4o-mini Triggered by @bot

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

0:0.07

1:0.08

2:0.08

3:0.07

4:0.08

5:0.08

6:0.09

7:0.08

8:0.08

9:0.08

10:0.08

11:0.07

Negative Logits

 Manufacturer

-3.17

 Painter

-3.00

Orn

-2.71

 Parliament

-2.70

 ornament

-2.60

 decorated

-2.58

 builder

-2.56

 simpler

-2.56

orn

-2.56

 parliament

-2.50

POSITIVE LOGITS

eph

3.04

ensen

2.91

iction

2.86

ictions

2.79

asc

2.66

 Jensen

2.56

 Neph

2.55

throp

2.52

WD

2.48

assment

2.47

Activations Density 0.000%

the word "as" used to indicate comparisons or explanations

No Comments

No Known Activations

the word "as" used to indicate comparisons or explanations

No Comments

No Known Activations