INDEX

Explanations

terms related to apology and responsibility

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

IFrame

Link

Not in Any Lists

No Comments

Head Attr Weights

0:0.01

1:0.08

2:0.18

3:0.06

4:0.01

5:0.06

6:0.07

7:0.12

8:0.08

9:0.12

10:0.10

11:0.06

Negative Logits

});

-1.08

quickShipAvailable

-0.94

PLIED

-0.92

})

-0.89

enario

-0.89

ioxide

-0.83

],

-0.82

unless

-0.82

peat

-0.79

 coasts

-0.79

POSITIVE LOGITS

 accordingly

1.12

 thereto

0.94

dur

0.91

arov

0.91

vik

0.87

 incrim

0.83

amaz

0.82

 scrut

0.81

Reloaded

0.81

 pron

0.80

Activations Density 0.327%

terms related to apology and responsibility

No Comments

No Known Activations