INDEX

Explanations

governmental or organizational terms and initiatives

oai_token-act-pair · gpt-3.5-turbo

terms related to plans, systems, or frameworks within a context of governance or organizations

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

 sqor

-0.72

sbm

-0.70

Ô

-0.70

ilaterally

-0.69

estern

-0.66

 Verse

-0.65

 Saying

-0.65

Interested

-0.64

?????-

-0.63

 Explan

-0.63

POSITIVE LOGITS

 itself

1.25

 relies

0.98

 attracts

0.97

 generates

0.96

 resembles

0.94

 persists

0.94

 thri

0.92

 lacks

0.92

 survives

0.91

 encourages

0.91

Activations Density 0.273%

governmental or organizational terms and initiatives

terms related to plans, systems, or frameworks within a context of governance or organizations

No Comments

No Known Activations