INDEX

Explanations

punctuation marks that indicate questioning

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

0:0.08

1:0.07

2:0.08

3:0.07

4:0.08

5:0.07

6:0.10

7:0.07

8:0.09

9:0.08

10:0.09

11:0.08

Negative Logits

ombies

-2.09

ukemia

-1.88

ospels

-1.70

alore

-1.69

gebra

-1.69

Dash

-1.64

zos

-1.63

vale

-1.63

fal

-1.62

lasses

-1.62

POSITIVE LOGITS

requ

1.71

consuming

1.59

�醒

1.57

 Photograph

1.54

 Humph

1.49

 permitting

1.45

ranging

1.45

AUD

1.44

ín

1.44

razil

1.43

Activations Density 0.000%

punctuation marks that indicate questioning

No Comments

No Known Activations