INDEX

Explanations

phrases related to personal relationships

oai_token-act-pair · gpt-3.5-turbo

expressions of gratitude and appreciation for relationships

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

velt

-0.79

edience

-0.67

{"

-0.65

oppers

-0.62

ogie

-0.62

leeve

-0.61

duc

-0.61

ogene

-0.61

ogle

-0.61

pletion

-0.60

POSITIVE LOGITS

 backstage

0.75

antically

0.68

 briefly

0.67

 cord

0.66

EngineDebug

0.65

 passionately

0.64

ãĥĥ

0.64

 exchanging

0.64

 peacefully

0.62

 onstage

0.61

Activations Density 0.265%

phrases related to personal relationships

expressions of gratitude and appreciation for relationships

No Comments

No Known Activations