INDEX

Explanations

specific positions or titles held by individuals in different contexts

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the presence of authoritative figures or authors

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

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

trap

-0.82

hazard

-0.71

 thereof

-0.71

ency

-0.68

hm

-0.68

VALUE

-0.68

liness

-0.67

complex

-0.67

leness

-0.67

Downloadha

-0.67

POSITIVE LOGITS

 founders

1.40

 pioneers

1.27

 architects

1.17

 recipients

1.13

 organizers

1.13

 authors

1.12

 beneficiaries

1.11

 principals

1.11

 founding

1.10

 plaintiffs

1.06

Activations Density 0.104%

specific positions or titles held by individuals in different contexts

phrases indicating the presence of authoritative figures or authors

No Comments

No Known Activations