INDEX

Explanations

mentions of behaviors or situations associated with extreme or irrational actions

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Lot

-0.64

LCS

-0.62

 silenced

-0.60

LECT

-0.60

âĢ¢âĢ¢

-0.58

NRS

-0.58

 Presbyterian

-0.57

ĵĺ

-0.57

PB

-0.57

 governed

-0.57

POSITIVE LOGITS

cheon

1.56

atics

1.46

atic

1.46

acy

1.34

atical

1.27

etic

1.15

acies

1.15

aris

1.14

acia

1.10

amic

1.10

Activations Density 0.042%

mentions of behaviors or situations associated with extreme or irrational actions

No Comments

No Known Activations