INDEX

Explanations

mentions of loved ones and relationships in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to loved ones

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

é¾

-0.78

ERO

-0.72

ulated

-0.70

ulhu

-0.69

illin

-0.64

erity

-0.64

ulation

-0.63

amphetamine

-0.63

IDER

-0.63

ipl

-0.62

POSITIVE LOGITS

 ones

1.06

 dearly

0.85

 pets

0.83

ometown

0.82

 uncond

0.79

 spouse

0.78

 nephew

0.78

 Ones

0.78

 memories

0.77

 loved

0.77

Activations Density 0.044%

mentions of loved ones and relationships in various contexts

references to loved ones

No Comments

No Known Activations