INDEX

Explanations

phrases indicating possibilities or speculation

oai_token-act-pair · gpt-3.5-turbo

modal verbs indicating uncertainty or possibility

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Toro

-0.67

 Patty

-0.65

 loved

-0.64

 Sunshine

-0.64

 Tome

-0.63

Viz

-0.62

OLD

-0.61

 Fare

-0.61

 Winged

-0.60

athing

-0.59

POSITIVE LOGITS

 occur

1.44

 signify

1.26

 preclude

1.22

 imply

1.22

 indicate

1.22

 originate

1.19

 contam

1.17

 arise

1.17

 correlate

1.17

 affect

1.16

Activations Density 0.187%

phrases indicating possibilities or speculation

modal verbs indicating uncertainty or possibility

No Comments

No Known Activations