INDEX

Explanations

phrases that inquire about different types or categories of things

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_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

0:0.04

1:0.01

2:0.08

3:0.07

4:0.07

5:0.03

6:0.14

7:0.09

8:0.06

9:0.05

10:0.05

11:0.25

Negative Logits

ogun

-2.01

EStream

-1.84

ocrat

-1.83

 cleans

-1.80

glomer

-1.76

obyl

-1.71

phthal

-1.70

orest

-1.65

itech

-1.62

iazep

-1.62

POSITIVE LOGITS

Filename

1.83

 autobiography

1.63

 bytes

1.54

 Nightmares

1.52

Translation

1.52

LOG

1.51

 signatures

1.51

 chapter

1.50

 Cipher

1.49

Discussion

1.45

Activations Density 0.003%

phrases that inquire about different types or categories of things

No Comments

No Known Activations