INDEX

Explanations

possessive pronouns followed by a word or phrase indicating ownership or relation

oai_token-act-pair · gpt-3.5-turbo

possessive pronouns and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Frie

-0.61

emale

-0.59

dn

-0.58

anamo

-0.57

ciation

-0.57

onica

-0.56

erb

-0.55

POSITIVE LOGITS

nces

0.64

ICE

0.62

ciples

0.59

OTOS

0.58

ELF

0.56

 flavorful

0.56

 VIDEOS

0.56

 sqor

0.56

 favourite

0.55

 sexuality

0.55

Activations Density 0.077%

possessive pronouns followed by a word or phrase indicating ownership or relation

possessive pronouns and related terms

No Comments

No Known Activations