INDEX

Explanations

occurrences of the word "by" indicating authorship or attribution

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_L3_Hcat_z_lr1.20e-03_l19.00e-01_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.3.attn.hook_z

Hook Layer

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.02

1:0.05

2:0.10

3:0.16

4:0.02

5:0.03

6:0.10

7:0.14

8:0.09

9:0.06

10:0.11

11:0.07

Negative Logits

azaki

-1.09

edi

-1.06

eve

-1.00

heit

-0.99

eon

-0.98

ioxide

-0.95

wana

-0.94

bia

-0.94

iasm

-0.93

mas

-0.92

POSITIVE LOGITS

.$

1.07

 visionary

1.06

 newcom

0.97

 extraord

0.96

$.

0.95

 srfAttach

0.94

 eleph

0.93

cloth

0.93

});

0.92

 taxpayers

0.92

Activations Density 0.060%

occurrences of the word "by" indicating authorship or attribution

No Comments

No Known Activations