INDEX

Explanations

words related to being impacted or influenced by a situation or event

oai_token-act-pair · gpt-3.5-turbo

references to entities or groups that are impacted by specific events or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ALK

-0.79

=-=-=-=-=-=-=-=-

-0.75

Blade

-0.75

ATT

-0.73

 Thou

-0.72

IDE

-0.71

ç«

-0.70

OVA

-0.68

 VIDEOS

-0.68

 Haku

-0.67

POSITIVE LOGITS

 affected

1.52

 impacted

1.36

 adversely

1.24

 effected

1.14

 unaffected

1.03

affected

0.98

 Affect

0.97

 affect

0.95

 negatively

0.93

 impact

0.86

Activations Density 0.012%

words related to being impacted or influenced by a situation or event

references to entities or groups that are impacted by specific events or conditions

No Comments

No Known Activations