INDEX

Explanations

the word "Things" occurring in sentences

oai_token-act-pair · gpt-3.5-turbo

references to "things" that are changing or evolving in context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Flavoring

-0.65

ECD

-0.64

asking

-0.59

iary

-0.58

owers

-0.58

iaries

-0.58

 Expend

-0.57

Joined

-0.57

doms

-0.57

bush

-0.57

POSITIVE LOGITS

 escalate

1.00

 calmed

0.96

 escalated

0.95

 deteriorated

0.93

 aren

0.92

 transpired

0.90

 weren

0.88

 went

0.87

 worsen

0.87

 changed

0.85

Activations Density 0.076%

the word "Things" occurring in sentences

references to "things" that are changing or evolving in context

No Comments

No Known Activations