INDEX

Explanations

phrases that indicate negative impacts or effects on various subjects

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

0:0.02

1:0.03

2:0.10

3:0.08

4:0.01

5:0.02

6:0.06

7:0.13

8:0.08

9:0.23

10:0.09

11:0.09

Negative Logits

displayText

-1.13

Untitled

-1.04

Haunted

-1.04

Downloadha

-1.02

used

-1.01

Reason

-1.01

��

-1.00

="#

-1.00

五

-0.99

deen

-0.99

POSITIVE LOGITS

 Flavoring

1.04

 Minor

1.01

 streng

1.00

 reconcil

1.00

 interoper

1.00

 uptake

0.97

OPS

0.96

 profitability

0.92

 behav

0.91

 Wasteland

0.90

Activations Density 0.017%

phrases that indicate negative impacts or effects on various subjects

No Comments

No Known Activations