INDEX

Explanations

the mention of exploitation in various contexts

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

0:0.06

1:0.02

2:0.15

3:0.05

4:0.07

5:0.26

6:0.05

7:0.01

8:0.17

9:0.04

10:0.04

11:0.03

Negative Logits

assetsadobe

-2.12

racuse

-1.76

TAIN

-1.67

SG

-1.55

ioxide

-1.53

arie

-1.49

CAST

-1.47

VERTIS

-1.44

[&

-1.44

quished

-1.42

POSITIVE LOGITS

qq

1.54

"""

1.54

\\

1.53

edi

1.53

**

1.48

*/

1.48

onga

1.45

 Hare

1.41

]"

1.40

ones

1.39

Activations Density 0.182%

the mention of exploitation in various contexts

No Comments

No Known Activations